Proceedings of the Zoological Society of London

O'b^^

PROCEEDINGS^

OF THE /

GENERAL MEETINaS FOS SCIENTIFIC BUSINESS

OP THE

ZOOLOGICAL SOCIETY

OF LONDON.

1903, vol. I.

(JANUARY— APRIL.)

PlllNTED FOR THE SOCIETY,
AND SOLD AT THEIR HOUSE IN HANOVEH-SQUARE.

LONDON :
MESSRS. LONGMANS, GREEN, AND CO.,

PATKU>."O.STF,U liUW.

LIST

OF THE

COUNCIL AND OFFICERS

OF THE

ZOOLOGICAL SOCIETY OE LONDON.

1903.

COUNCIL.

{Elected April 29th, 1903.)
His Grace Tue Duke op Bedford, K,G., President.

William T. Blanfoed, Esq.,

I.Ij.D.,F.B..S., Vice-President.
George A. Boulexgee, Esq.,

F.R.S., Vice-President.
William E. de Winton, Esq.,

Acting -Superi7itendent of the

Gardens.
Herbert Deuce, Esq., F.L.S.
Chaeles Drummond, Esq.,

Treasurer.
Charles H. Gatty, Esq., LL.D.
Feedeeick Gillett, Esq.
F. DuCane Godbian, Esq.,

D.C.L.,F.Il.S., Fice-Presiffeii.
Albert Gunthee, Esq., M.D.,

Ph.D., F.R.S., Vice-President.

Oapt. The Marquis of Hamil-
ton, M.P.

Prof. George B. Howes, D.Sc,
LL.D., F.R.S., Vice-President.

Lt.-Ool. L. Howard Ieby.

Sir Edmund G. Loder, Bt.

E. G. B. Meade- Waldo, Esq.

P. Chalmers Mitchell, Esq.,
M.A., D.Sc, Secretary.

Sir Thomas Paine.

E. Lort Phillips, Esq.

David Sharp, Esq., M.D.,
F.R.S.

Oldfield Thomas, Esq., F.R.S.

Henry Woodward, Esq., LL.D.,
F.R.S. , Vice-President.

PRINCIPAL OFFICERS.

P. Chalmers Mitchell, Esq., M.A., D.Sc, Secretary.
W. E. DE Winton, Esq., Acting-Superintendent of the

Garde7is.
Frank E. Beddard, Esq., M.A., F.R.S., Prosector.
Me. F. H. Waterhouse, Librarian.
Mr. John Barrow, Accoicntant.
Mr. W. H. Cole, C?tief Clerk.

Mr. George Arthur Doubleday, Clerk of Publications.
Me. Arthur Thomson, Assistant Superintendent of the

Gardens.

LIST OF COiNTENTS.

January 20, 1903.

Page

The Secretary. Ropoit on the Additions to the Society's

Menagerie in December 1902 1

Mr. P. L. Sclater. On the Zebra-and-Pony Hybrid living

in the Society's Menagerie 1

Mr. J. S. Budgett, M.A., F.Z.S. Account of his recent
Journey to Uganda in search of the Okapi and
Polyjyterus 2

1. Note on the Spiracles of Polypterus. By J. S. Budgett,

M.A., F.Z.S 10

2. On the Brains of N'asalis larvahts and of some otlier

Old World Primates. By Frank E. Beddard, F.R.S.

&c 12

3. On the Fishes collected by Mr. G. L. Bates in Southern

Cameroon. By G. A. Boulenger, F.R.S. , V.P.Z.S.
(Plates I.-V.) 21

4. On the Anatomy of the Gcphyrean PJiascolosoma teres,

n. sp. By W. K. IIuttox, M.A., M.B., Senior Demon-
strator of Anatomy in the University of Glasgow.
(Plates VI.-VIII.) 29

5. On Potamon {Potamonautes) latidactylum, a new Ficsh-

watoi' Crali from Upper Guinea. By Dr. J. G. de Man,

of lerseke, Holland. (Plato IX.) 41

a 2

IV

Page

6. On a new " Bird's-dung " Spider from Ceylon. By R. I.

PococK, F.Z.S., and the Hon. IST. 0. Rothschild, B.A.,
F.KS., F.Z.S. (Plate X.) 48

7. On the Crustaceans of the Genera Petalidium and Ser-

gestes from the ' Challenger,' with an Account of
Luminous Organs in Sergestes chcdlengeri, n. sp. By
Dr. H. J. Hansen (Copenhagen). (Plates XI. & XII.) 52

February 3, 1903.

The Secretary. Report on the Additions to the Society's

Menagerie in January 1903 • — 79

1. Notes on the Hair-Slope of four Typical Mammals. By

Walter Kidd, F.Z.S 79

2. A Prodromus of the Snakes hitherto recorded from China,

Japan, and the Loo Choo Islands ; with some Notes.
By Captain F. Wall, Indian Medical Service 84

3. Note on the Wild Sheep of the Kopet-Dagh. By R.

Lydekker 102

4. On new Parasitic Copepoda from Zanzibar and East

Africa, collected by Mr. Cyril Crossland, B.A., B.Sc.
By Staff-Surgeon P. W. Bassett-Smith, R.N., F.Z.S. ... 104

5. On the Original Home of the Tiger. By Col. C. E.

Stewart, C.B., C.M.G., CLE 109

6. On the Mode of Copulation of the Indian Elephant. By

H. Slade, Conservator of Forests, Maymyo, Burma ... Ill

7. On the Coelenterata collected by Mr. C. Crossland in

Zanzibar. — I. Ceratella minima, n. sp. By Sydney J.
HiCKSON, M.A., F.R.S., F.Z.S., Beyer Professor of
Zoology in the Owens College, Manchester. (Plate
XIIL) 113

8. Contributions to our Knowledge of the Plankton of the

Faeroe Channel. — No. YIII. By G. Herbert Fowler,
B.A., Ph.D., F.Z.S 117

9. On the Present Condition and Habits of the Elk in

Norway. By H. J. Elwes, F.R.S 133

V

Fel)riiary 17, 1903.

Page
Mr, R. E. Holding. Exhibition of, and remarks upon, some

Skulls of Mammals showing abnormal dentition 151

1. Oa some new Species of Spiders belonging to the Families

Flsauridce and Senoculidce ; with Cliaracters of a new
Genus. By Frederick Pickard-Cambridgb, B.A.,
F.Z.S. (Plates Xiy. & XY.) 15i

2. On the Marine Fauna of Zanzibar and British East Africa,

from Collections made by Cyril Crossland in the Years
lyul and 1902.— PolycluBta. Pai-t 1. By Cvmul Cross-
LAXD, B.A., B.Sc. (Plates X.\L Sc XYII.) 169

3. On the Axis, Atlas, and Proatlas in the Higher Tlierio-

donts. By R. Broom, M.D., B.Sc, C.M.Z.S. (Plate
XYIII.) 177

4. A Revision of the Fishes of the Genus Triacanthus. By

0. Tate Regan, B.A 180

J5. On the Geographical Yariations of the Sand-Yiper, Vipera

aumwdytes. By G. A. Boulenger, F.R.S., Y.P.Z.S. ... 185

6. Motes on the Habits of the Hoolock. By Geo. Candler,

M.B.Cantab 187

March 3, 1903.

The Secretary. Report on t]u^ Additions to tlie Society's

Menagerie in February 1 903 191

The Seci-etaiy. Extract from a letter from Major C. Del me
RadcliU'e concerning skins of a Monkey and an Otter
from Uganda 191

Mr. F. E. Beddard, F.R.S. Exhibition of a mounted skin

of the Greater Bird of Paradise 1 92

Ml'. J. L. Bonhote, F.Z.S. Exhibition of, and remarks upon,
a photogniph of Ek'pli.-ints showing considei'able gi-owth
oi hair, and note upon the Sanskrit name of the Tiger... 192

Prof. F. Jeffrey Bell, F.Z.S. Exhibition of, and remarks

upon, a Holotburian of tlu' Genus Actinopi/ga 192

1. On a new Species of Pigmy Antelope of the Genus jS'^eo-
tra<jus from the Cameroous District, W. Africji. By
W. E. DE WiisToN, F.Z.S. (Plate XIX.) '.. 192

yi

2. On the Land Operculate Mollusca collected during the

" Skeat Expedition " to the Malay Peninsula in 1899-
1900. By E. K. Sykes, F.Z.S. (Plate XX.) 194

3. The Significance of the Callosities on the Limbs of the

Equidce^ By R. Ltdekker, F.Z.S 199

4. Note on some Remains of Struthio karatheodoris Maj.

of the Island of Samos. By Rudolf Martin, of Basel
University 203

5. On a new Genus and two new Species of Earthworms

of the Family Eudrilidce, with some Notes upon other
African Oligochseta. By Frank E. Beddard, M.A,,
F.R.S., F.2.S. 210

March 17, 1903.

Prof. Newton, F.R.S. Exhibition of, and remarks upon,
photographs of the White Rhinoceros, taken by Mr.
C. R. Saunders, C3I.G., in Zululand 222

Mr. Oldfield Thomas, F.R.S. Exhibition of a skin and
description of a new species of Monkey, Ehinopithecus
hrelichi. (Plate XXL) 224

Mr. Oldfield Thomas, F.R.S. Exhibition of specimens and
description of a new species of Dixiker, Gephalophus
ignifer 225

1. Observations and Experiments on Japanese Long- tailed

Fowls. By J. T. OuNNimHAM, M.A., F.Z.S. 227

2. On some Nudibranchs from East Africa and Zanzibar. —

Part 11. By Sir 0. Eliot, K.O.M.G., H.M. Com-
missioner for the East Africa Protectorate, F.Z.S. 250

3. Contributions to the Osteology of Birds. — Part VL

Cucidiformes. By W. P. Pycraft, F.Z.S., AX.S.
(Plate XXII.) 258

April 21, 1903.

The Secretary. Report on the Additions to the Society ''s

Menagerie in March 1903 292

L Linnoeus and Hunter on Feather-Tracts. By Henry

Scherren, F.Z.S 292

Vll

Pago

2. On some Mammals collected by Cajit. II. N. Dunn,

R.A.M.C, in the Soudan. By Oldfielu Thomas,
F.R.S., F.Z.S 294

3. On a Collection of Turbellaria Polycladida from the Straits

of Malacca. (Skeat Expedition, 1899-1900.) By F. F.
Laidlaav, B.A.Cantab., Assistant Lecturer and Demon-
strator in the Owens College. (Plate XXIII.) 301

4. On the Phylogenetic Cause of the Transposition of the

Testes in Mammalia : with Remai'ks on the Evolution
of the Diaphragm and the Metanephric Kidney. By
W. Woodland, Univei'sity College, London 319

5. On the Geographical Distribution of Spidci's of the Oi-der

Mygalomorplue. By R. I. Pocock, F.Z.S 340

EERATDM.

'age 110, ]ine 29, for Archipelago, which read Archipelago which.

A L r II A B B T I C A L LIST

CONTRIBUTORS,

With References to the several Articles contributed by ectch.

Puge
Bassett-Smith, Staff-Surgeon P. W., R.N., F.Z.S.

On new Parasitic Oopepoda from Zanzibar and East
Africa, collected by Mr. Cyril Orossland, B.A., B.Sc 104

Beddard, Frank E., M.A., F.Pt.S., Prosector to the Society.

On the Brains of Nasalis larvatus and of some other
Old World Primates 12

Exhibition of a mounted skin of the Greater Bird of
Paradise 192

On a new Genus and two new Species of Earthworms
of the Family Euclrilidte, with some Notes upon other
African Oligochreta 210

Bell, Prof. F. Jeffrey, M.A., F.Z.S.

Exhibition of, and remarks upon, a Ilolothurian of the
Genus Aciinopi/ga 192

X

Page
BoNHOTE, J. Lewis, M.A., F.Z.S.

Exhibition of, and remarks upon, a photograph of
Elephants showing considerable growth of hair, and note
upon the Sanskrit name of the Tiger 192

BouLENGER, George Albert, F.R.S., V.P.Z.S,

On the Fishes collected by Mr. G. L. Bates in Southern
Cameroon. (Plates I.-V.) 21

On the Geographical Variations of the Sand-Yiper,
Vipera ammodytes 185

Broom, Robert, M.D., B.Sc, C.M.Z.S.

On the Axis, Atlas, and Proatlas in the Higher Therio-
donts. (Plate XVIII.) 177

BUDGETT, J. S., M.A., F.Z.S.

Account of his recent Journey to Uganda in search of
the Okapi and Polypterus 2

Note on the Spiracles of Polypterus 10

Cambridge, F. Pickard. See Pickard-Cambridge, F.

Candler, George, M.B. Cantab.

Notes on the Habits of the Hoolock 187

Crossland, Cyril, B.A., B.Sc.

On the Marine Fauna of Zanzibar and British East
Africa, from Collections made by Cyril Crossland in the
Years 1901 and 1902.— Polychseta. Part I. (Plates
XYI. &XVII.) 169

Cunningham, J. T., M.A., F.Z.S.

Observations and Experiments on Japanese Long-tailed
Fowls 227

De Man, Dr. J. G., lerseke, Holland.

On Potamon (Potamoncmtes) latidactylimi, a new Fresh-
water Crab from Upper Guinea. (Plate IX. ) 41

Page

De Wintox, William E., Acting- Superintendent of the
Society's Gai-dens.

On a new Species of Pigmy Antelope of the Genus
Reotragus from the Cameroons District, W. Africa.
(Plate XIX.) 11)2

Eliot, Sir Charles, K.O.M.G., H.M. Commissioner for
the East Afiica Protectorate, F.Z.S.
On some Nudibranchs from East Africa and Zanzibar. —
Part II 250

Elwes, H. J., F.R.S., F.Z.S.

On the Present Condition and Habits of the Elk in
Norway 133

Fowler, G. Herbert, B.A., Ph.D., F.Z.S.

Contributions to our Knowledge of the Plankton of the
Faeroe Channel.— No. VIII 117

Hansen, Dr. H. J., Copenhagen.

On the Crustaceans of the Genera Feicdidium and
Sergestes from the ' Challenger,' with an Account of
Luminous Organs in Sergestes challengeri, n. sp. (Plates
XI. &XII.) 52

HiCKSON, Sydney J., M.A., F.R.S., F.Z.S., Beyer Professor
of Zoology in the Owens College, Manchester.

On the Coelenterata collected by Mr. C. Crossland in
Zanzibar. — I. Ceratella minima, n. s\). (Plate XIII.) ... 113

Holding, R. E.

Exhibition of, and remarks ui^on, some Skulls of
Mammals showing abnormal dentition 151

HuTTON, W. K., M.A., M.B., Senior Demonstrator of
Anatomy in the University of Glasgow.

On the Anatomy of the Gephyrean rhascolosoma teres,
n. sp. (Plates VI. -VIII.) 29

Xll

Page'

KiDD, Walter, M.D., F.Z.S.

Notes on the Hair- Slope of four Typical Mammals ... 79

Laidlaw, F. F., B.A.Oantab., Assistant Lecturer and
Demonstrator in the Owens College.

On a Collection of Turbellaria Polycladida from the
Straits of Malacca. (Skeat Expedition, 1899-1900.)
(Plate XXIII.) 301

Lydekker, Richard, B.A., F.R.S., F.Z.S.

Note on the Wild Sheep of tlie Kopet-Dagh • 102

The Significance of the Callosities on the Limbs of the
Equidce 199

Martin, Rudolf, of Basel University.

Note on some Remains of Struthio karatheodoris Maj.
of the Island of Samos 203

Newton, Prof. Alfred, M.A., F.R.S., F.Z.S.

Exhibition of, and remarks upon, photographs of the
White Rhinoceros, taken by Mr. C. R. Saimders, C.M.G.,
inZululand 222

Pick ard- Cambridge, Frederick, B.A., F.Z.S.

On some new Species of Spiders belonging to the
Families Pisauridm and Senoculidce ; with Characters of
a new Genus. (Plates XIV. & XV.) 151

PococK, R. I., F.Z.S.

On the Geographical Distribution of Spiders of the
Order Mygalomorphee 340

PococK, R. I., F.Z.S., and Rothschild, Hon. N. C, B.A.,
F.E.S., F.Z.S.

On a new "Bird's-dung" Spider from Ceylon. (Plate X.) 48

Xlll

Pago
Pycraft, William Plane, F.Z.S., A.L.S.

Contributions to the Osteology of Birds. — Part Yl.
Cacidlformes. (Plate XXII.) 208

Regan, C. Tate, B.A., British Museum, Natural History.

A Revision of the Fi.shes of the Genus Triacanthus ... 180

RoTiiscniLD, Hon. N. C, B.A., F.E.S., F.Z.S., and Pocock,
R. I., F.Z.S.

On a new " Bird's-dung" Spider from Ceylon. (Plate X.) 48

ScHERREN, Henry, F.Z.S.

Linnreus and Hunter on Feather-Tracts 292

Sclater, Philip Lutley, M.A., D.Sc., Ph.D., F.R.S.

(Secretary to the Society till January 20th, 1903).

Report on the Additions to the Society's Menagerie in
December 1902 1

On the Zebra-and-Pony hybrid living in the Society's
Menagerie 1

Sclater, William Lutley, M.A. (Secretary to the Society
from January 21st to April 29th, 1903).

Report on the Additions to the Society's Menagerie in
January 1903 79

Report on the Additions to the Society's Menagerie in
February 1903 1^1

Extract from a letter from IMajor 0. Delrac Radclifle
concerning skins of a Monkey and an Otter fi'om Uganda. 191

Report on the Additions to the Society's Menagerie in
March 1903 292

Slade, H., Conservator of Forests, Maymyo, Burma.

On the Mode of Copulation of the In<lian Elephant ... HI

XIV

Page
Stewart, Col. 0. E., C.B., C.M.G., CLE.

On the Original Home of the Tiger 109

Sykes, Ernest Ruthven, F.Z.S.

On the Land Operculate MoUusca collected during the
' Skeat Expedition ' to the Malay Peninsula in 1899-1900.
(Plate XX.) 194

Thomas, Oldfield, F.R.S., F.Z.S.

Exhibition of a skin and description of a new species
of Monkey, Rhinopithecus hrelichi. (Plate XXL) 224

Exhibition of specimens and description of a new species
of Duiker, Gephalojjhics ignifer 225

On some Mammals collected by Capt. H. N, Dunn,
R.A.M.O., in the Soudan 294

Wall, Captain F., Indian Medical Service.

A Prodromus of the Snakes hitherto recorded from
China, Japan, and the Loo Ohoo Islands ; with some Notes. 84

Woodland, W., University College, London.

On the Phylogenetic Cause of the Transposition of the
Testes in Mammalia : with Remarks on the Evolution , of
the Diaphragm and the Metanephric Kidney 319

LIST OF PLATES.
1903.— Vol.. I.

21

Plate rage

1. 1. Alestes intennedms. '1. Alestes opisthotauid. 3. Am-^^

philiua lonf/irostris

II. 1. Lubeo anncctens. 2. Barhus tceniurus

III. 1. Barbels jx'ogenys. 2. Barbus batesii

IV. Microsyjiodontis batesii |

V. 1. Anabas pleurosti(/)iia. 2. Mastacembelus schcteri . . . . )

YI.i

MI. y Anatomy and Histology of Phascolosoma teres 29

VIII. i
IX. Figs 1-6. Potamon {Potamonnutes) laiidacii/luin. Figs.

7-9. Potatnon [Potanwnaute») africanum 41

X. PhrynaravJine rothschildi 48

XI. 'F'l^.l. Petalidiumfoliaceum. 2. Petalidium ]ni\. ^. Scr- \

gestes profu7ulus. 4. S. prehensilis. 5. S. kroyeri. I ^..^

6. S. similis \

XII. Fig. 1. Sergestes arcticus. 2. jS. challerigeri J

XIII. Ceratella minima 113

XIV. Spiders of the Families Pisauridae and Seuoculidae .... ( -,-•,
XV. Spidera of the Family Pitiuurida2 I

XVI. Figs. 1, 2, 3, 4, 7, 8. Phyllochcctopterus elioti. Figs. 5, j

6, 9. P. pictus i 109

XVII. Phylluchcetopterus elioti J

XVIII. Axis, Atla^i, and Proatlas of Gomplioynathua and Tri-

rachudon 177

XIX. Ncotragus batesi , 192

XX. Mollusca from the Malay Peninsula 194

XXI. lihinopithecus brelichi 224

X.XII. Osteology of the Cuculiformcs 2o8

XXIII. Polyclads from the Straits of Malacca 301

LIST OF TEXT-FIGURES.
1903.— Vol. L

Page

1. HyLrid between Equus burchelli, c?, and E. caballus, $ 2

2. Upper surface of cerebral hemispheres of Nasalis 13

3. Another brain of the same species 13

4. Longitudinal median section of brain of Nasalis 15

5. Upper surface of cerebral hemispheres of Colobtis guereza 17

6. Longitudinal median section of brain of Colobus guereza 17

7. Lateral view of brain of Cynopithecus 7iiger 19

8. Longitudinal median section of brain of Cynopithecus niger .... 19

9. Tentacular wreath of («) Phascolosoma and (b) Phyniosoma .... 35

10. Front view of skull and horns of adult Ram of the Kopet-Dagh

Urial 103

11. Chondrocarpus reticulosus and Chondrocarpus sp 105

12. Ventriculina crosslandi 107

13..

14.

15. VTelsons of larvse attributed to Thysanoessa longicaudata ...... 131

16.
17.^

18. Horns of fully adult Elk from Trondhjem, Norway 134

19. Horns of fully adult Elk killed in Bjorndal 134

20. Horns of a full-grown but not adult Elk, killed in Upper

Namdalen 135

21. Horns of young Elk, supposed to be 3 years old 135

22. Tracing of shed horn picked up at Mo 142

23. Horns of a large Elk from Lithuania, in Branicki Museum .... 143

24. Horns of a very old Elk showing degeneration 147

25. Tracing of a cast horn found at Solem in Bangdal by Capt.

Ferrand 148

26. One-horned Elk in Ipswich Museum 149

27. Side views of heads of («) Vipeirc ammodytes f. typica and {b)

var. meridionalis 186

28. Front view of end of snout, showing the lepidosis 186

XVll

Pa-re

:2!). Slcnll of Xcntrarpia hnfcsii l'.).'>

•'O. liigiit ieinur of Striithio havaiJieoclorig. Caudal view 204

'31. Txv^ht iemnx oi Strutliio karatheodoris. Rostral view 204

32. Pelvis of Struthio karatheodoris. Ventral view 20."»

33. Pelvis of Struthio karathcodorls. Dorsal view 20(5

34. Lateral view of the pelvis of Struthio karathcodori^ 207

3o. Di.^section of Stuhhnnnniu michaelmmi 211

30, Ventral surface of Bettonia lat/ariensii 214

37. Termination of the male ellerent apparatus of Ilettoniit. liujari-

ensis 215

38. Spennatophore of rurcudrilus ,sp 220

39. Eecently-killed lihinoceros siinus, adult J 223

40. Itecently-killed Rhinoceros simus, adult c? 224

41. Japanese Long-tailed Fowl. Cock A, photographed April 1903. 240

42. Japanese Long-tailed Fowl. Cock Bjphotograplied April 1903. 241

43. Left side view of stiu-uum and shoulder-girdle of CuchIhs

cdnorus 277

44. Left side view of sternum and shuulder-girdle of Voua rcyuaudi . 277

45. Dorsal aspect of the pelvis of Cacomnntis merulinus 280

40. Dorsal aspect of the pelvis of Geococcyx mcriccmus 282

47. Side view of same 282

48. Ventral aspect of same 282

40. Topographical diagram showing feather-tracts and bare spaces

in schematic bird , 294

50. Eye-spots of Notoplana evansii 303

51. Male orji'ans of Notoplana ei-am^ii 304

52. Genital apparatus of Leptoplana malaijana 300

53. " Brain-eyes " of Scmonia peinnif/cnais 30i)

54. Genital apparatus of Bcrrieiidalia anomala 311

55. Eye-spots of Latocest.es arr/us 313

50. Eye-spots of Prosthiostovm/n pal/idmn 317

57. Diagram illustrating Testis Descent 334

58. Map to illustrate the Geographical Distribution of tlie Macro-

thelean genera of Dipluridca 344

59. Map to illustrate the Geographical Distribution of the Atypidce,

Brachyhothriidcr, and Mecicohotliriidrf 347

60. Map to illustrate tlie Geographical Distribution of the C'fenizidcs

and Miyidce 351

61. Map to illustrate the Gei>graphical Distribution of tin; Avicu-

lariidce 355

Proc. Zuol. Sue. — 1903, Vol. I.

LIST OF NEW GENERIC TERMS

PROPOSED m THE PEESENT VOLUME (1903, vol. I.).

P^ge

Acolischnus (Aracbn.) 362

Asthenoccros (Vermes) 315

Bergendalia (Vermes) 310

Ceratophyllidia (Moll.) 250

Chondrocarpiis (Copep.) 104

Microsynodontis (Pisces) 26

Notoplana (Vermes) 302

Paradossenus (Arachn.) 153, 155

Pleurophyllidiella (Moll.) 251

Ventriculina (Copep.) 106

PROCEEDINGS

GENERAL MEETINGS FOR SCIENTIFIC lUJSINKSS

OF THE

ZOOLOGICAL SOCIETY OF LONDON
1903, Vol. I. (January to April).

January 20, 1903.

Prof. G. B. Howes, D.Sc, LL.D., F.R.S., Vice-President,
in the Chair.

The Secretary read the following report on the additions to the
Society's Menagerie during the month of December 1902 : —

The I'egistered additions to the Society's Menngerie dui'ing the
month of December 1902 were 30 in number. Of these 15 were
acquired by presentation, 1 was born in the Gardens, and 10 were
received on deposit and 4 in exchange. The total number of
departui-es dui-ing the same period, Ijy death and removals,
was 141.

Amongst the additions worthy of notice are two very fine speci-
mens of the One-wattled CtiHf^ov/iuy (Oasaarms uniapjiendicidatus),
from New Guinea, deposited by the Hon. Walter Rothscliihl,
M.P., F.Z.S., on Dec. 30th.

Mr. P. L. Sclater read an exti'act from a letter from Messrs.
Stagmann, Esselen & Roos, of Pretoi'ia, dated Dec. 8th, 1902, and
addressed to Major W. H. Birkbeck of the Remount Department,
Johaiinesberg, fi'om whicli it appeared tJiat the hylnid Zebra,
placed inider the Society's cai-e liy tlie King on July IDtli. 1902

Proc. Zool. Soc— 1903, Vol. I. No. I. 1

ME. J. S. BUDGETT ON

[Jan. 20,

(see P. Z. S. 1902, vol. ii. p. 225), was the progeny of a male
Zebra and a pony mare. Major Birkbeck added the following

Text-fig. 1.

Hybrid between Eqtms hurclielU, $ , and IS. cahallus, !j? .

particulars: — The hybrid was rising 12 months in July 1900
when he came in. He went home with a pony of which he was
very fond, and if he gets troublesome the company of a quiet
pony will pacify him at once. We gelded him, thinking Her
Majesty might use him, and a stallion hybrid is always a terror.
The late owner Erasmus has tried to claim compensation, but as
the beast was a fair capture he has no case at all. At the
surrender the Boers kept any horses of ours they had and we, of
course, keep theirs.

Mr. J. S. Budgett, M.A., F.Z.S., gave an account of his recent
journey to Uganda and return by the Nile, which was illustrated
by a large series of photographs taken by him as he went along.

Mr. Budgett made the following remarks : —

The special object of my journey to Uganda was to continue
investigations on the life-history of African fishes and especially

1903.] niS JOURNEY TO UGANDA. 3

of Polypteriis. Having made previous attempts to solve this
problem iimler certain eonditions in a confined arcix, it was
thought that by ol)serving it uiider varied conditions of latitude
and altitude, new light might be brought to bear upon it.

Fi'om Uganda it was possible either to work down the Congo
from its source or to return northwards down the Nile.

On my way to Uganda, and in Uganda., I gathered what infor-
mation I could about the two routes. The southern end of the
Nile Valley really lies at the foot of Ruwenzori, while a short
journey westwards from this point would bring one into the he.ul-
w:i.ters of the Congo.

A special inducement to t;dve the Congo route was that infor-
mati(»n might in this joui'uey be obtaineil a,l)o>it the new animal
Okapiajnhnstnni, and other interesting forms, Ix^lieved to exist in
the )Semliki Forest.

From information gathered in Uganda, it was clear that it was
useless to hope to meet with Okapi in British territory, and,
moreover, I here learned that the Belgians had found the Okapi
in large numbers in the Welle country.

I found also, what had been veiy difficult to learn before
leaving home, that the season of the rains and the breeding of
Polypteriis were considerably earlier at the source of the Nile than
they wei-e further northwards ; that the Semliki Valley was a,
most inconvenient place at which to mn.ke a permanent camp
by the river-banks, owing chiefly to the scarcity of food; and
that only one species of Polyptei'its was to be found there, while
at least three species were to be found in the Nile fait.her to the
north.

The difficulty of taking delicate appa,ratus through the Congo
Forest to the upper watei-s of the Congo was incompara.1)ly
greater than to the upper Nile.

And, lastly, it was to be borne in mind that the time of year at
which one might hope to be successful in the main object was
that at which it was well nigh impossible to do much in the way
of collecting the higher Vertebrata which might be supposed to be
of interest in the Semliki Valley ; for at this time the grass is so
high that moving away from beaten tracks is almost impossible,
while anything smaller than giraffi^ or elephant is seldom seen.
Bearing these fa,cts in mind, I had little hesitation in (h^ciding to
work down the Nile, sti'iking it at a point fa,rt]ier northwards
than the iSemliki River, in order to take advantage of the lateness
of the sea.son in that region. Accordingly, having fitted o>it my
safari or cai-avan at Entebbe, I started for Butya,ba, on the west
coast of Lake Albert, on July 10th. I had the advantage of
starting thus along a. good roa,d which had just been cleared for
the greater part of the way, and along which the rest-houses had
been repaired for the convenience of the Commissioner and
Consul-General of British East Africa, Sir Charles Eliot, who
has recently made such a. i-emarknldy rapid journey from Entebl)e
to Gondokoro, the frontier station of Uganda on the Upper Nile.

1*

4 MR. J. S. BUDGETT ON [Jan. 20,

At Kampala I diverged, however, for a few days along the old
road to Massindi to the eastward.

I had with me at the start 50 men and boys and my bicycle.
So many books have been written on Uganda, that there is
little need to describe the scenery of these tropical highlands,
especially as Sir Harry Johnston's wonderfully complete book is
now in everybody's hands.

Shortly, one may say that, on going northwards from Lake
Victoria, forest is hardly seen after leaving Kampala. We passed
day after day through almost endless elephant-grass, with palm-
groves and papyrus-swamps in the lower parts. The hills are
clothed with clumps and patches of acacia and euphorbias,
while their summits are very frequently covered with huge
granite boulders. There were thunderstorms and rain every
afternoon, and for the first few days I saw little in the way of
animal life : occasionally a Civet cat would cross the path, while
overhead Hornbills and Plantain-eaters of various species were
common. In the valley of the Maangia for a time we were rid -
of the everlasting elephant-grass, and here Gohus thomasi and
Zebra were plentiful. The Cohus thomasi of this region is some-
what different to that met with in the valley of the Nile, the
horns having a wider curve and being stouter and of a lighter
colour, while the animal itself is of a larger build and has more
brilliant markings.

The Maangia River flows northward through wide undulating
plains, covered at this time with hay-like grasses upwards of five
feet in height, and dotted over with very fine acacias, of a cedar-
of-Lebanon appearance and of richest deep-green colour. The
grasses and bushes of the roadside teem with bird-life : as we
rode along, the little Vidua principalis, with his dozen sombre
wives, was a constant companion, flitting just ahead of us for
a mile or more along the road. Likewise the Common Shrike
of these parts [Lanius excuhitoritis) has exactly the same habit
of driving along in flocks with a caravan as the Corvinella
corvina of the West Coast. In the marshy parts, Scopus umbretta
was often seen.

Then, leaving the plain, we struck up over the Bukamva hills,
and at the crest dived into the dense grass at the side of the road,
to travel for four weary hours over a wretched and little-used
track, often obliterated by the tramp of elephants, and where it was
quite impossible to make any progress with the bicycle. Then,
descending rapidly by swampy valleys and thick jungle, we came
suddenly into the new road to Hoima and Butyaba. From here to
Hoima the road crossed the steep hills which form the boundary
between Uganda and Unyoro, passing by abundant plantations of
bananas and through many a lovely valley, at the bottom of which
a stream ran through the richest vegetation, the banks carpeted
with Cannas, winter cherries, and hemlock, while overhead were
many Pterocarptos-trees with blossoms like the Alamanda flower.
As I did not care to get too far from my safari, I would often

1903.] HIS JOURNEY TO UGANDA. 5

bicycle on for an hoiu-, and then sit down in one of these shady
spots, and watch the Mouse-])irds han<,ang Hke acrobats in attitudes
most quaint, and Sun-birds darting in and out of the great red
blossoms of the Spathodea, while often a noisy flock of Prionops
plumatus passed hurriedly along. In the moi-e open parts I often
saw several pnu's of Ground Hornbills, and each time one had a blue
throat, the other a red one. I can but think that this is a sexual
character, though which is the female and which is the male
I was unable to detei-mine. Other bii-ds seen were Pceocephalus
meyeri, a species of Mct^'onyx, Irrisor erythrorhynchits, Hirundo
7'iostica, and the handsome Snipe Rhyiichciia capensis.

A very amusing bii-d that 1 watched was Eryihropygia ruficauda,
which is most assiduous in its courting of the female, spreading
its tail before her like a fan.

In the more shady parts one might often see the butterflies, as
I reniendier seeing them in the forests of Paraguay, covering the
ground with large patches of coloui-, in flocks according to their
species.

Shortly before i"e;iching Hoima the i-iver Kaf u is ci'ossed ;
here a netwoi-k of papyi-us-swamjis with good causeways over
them abound in duck, geese, and kingfishers of vaiious kinds.
Then, winding upwards, a high point is reached fi'om which the
village of Hoima is seen, and in the veiy fai- distance one can
make out the Blue Moimtains on the other .side of Lake Albert.
At this high point I saw a charming little Widow- bird [Vidua
hypocherina). Lions round this part aie plentiful and somewhat
dangerous, as they usually ai-e in countries where game is not
abvmdant. At Hoima 1 heard that several natives had recently
been carried oS* by the Lions.

From here, two days' march through the so-called Budonga
Forest brings one to the shores of Lake Albert. This Budonga
Forest is nothing more than rather heavily- wooded scrub. It is
true that in the ravines and gorges there ai-e strips of real forest,
but it is not in any way comparable with tlie I'eal forest of the
tropics, where the sky can scarcely be seen.

This Budonga woodland teems with herds of Elephant — I
myself calculated that there were over 200 in one herd which
we came across. Some of the males had enormous tusks, and
these big fellows seem to keep slightly aloof fi'oin the rest of the
herd. I knew that we were quite close to this hei'd, as there were
great roadways through the jungle with cpiite fresh, smoking
dung ; and here I first notice<l what stiuck me many times
subsequently, that when elephant-dung falls on a pathway or
clearing, there within half an hoiu- you will constantly find,
heaped up all round the <lung, the earth-workings of a shrew or
mole. What is it the shrew seeks in the dung? Is it the fly-
larva? that have been blown upon the dung, or is it the dung
itself ? Frequently elej)hants in these parts appear of a bright
red colour, having covered their bodies with the dust of crushed-
vip termite hills.

6 MR. J. s. BUDGETT ON • [Jan. 20,

On July 29th I looked from a high point on the road on to
Lake Albeit, a vast sheet of glistening water, 1000 feet below,
bordei-ed on this side with level plains of park-land, broken hei'e
and there by lagoons and swamps, where I was to try first for the
Polyptertbs.

Of the results of my Polypterus work during this journey I
shall say nothing hei-e ; sufiice it that I stayed down by the
lake-side from July 30th to August 15th, trapping, netting, and
shooting. During this time the fishes most abundantly met with
were Hyclrocyon forskalii, Alestes haremose^ Dislichodus niloiicus,
Laheo hosei, Bagriis hayad, Eutrojims niloticus, Synodontis nicjritus,
Tilcqna nilotica, also very large specimens of a Citharinus.

Lates niloticihs is fi'equently caught by the natives here 5 and
6 feet long, usually with the spear. ProtojJterus and Polypterus
were both obtained hei-e. The River- Toi'toise {Trionyx triunguis),
28 inches in length, and veiy lai'ge specimens of Rana occijntalis
were also common here.

The common Antelopes were Cobus thomasi, Cobus defassa,
Tragelaphios scriptus, Oiibi, and Cejjhcdophtcs ceqitaiorialis. Down
on these lake- side flats the avifauna difters in a marked mannei-
from that in the highlands. Hyjyhantornis cucullatus was now
building in hundreds in the water-side bushes ; Laniai-msbarbarus
a,nd Telep)lionus senegaltis in the low bushes, with Merops albicollis,
LairiproGoli'iis p>ur'inireus^ Pyrotnelana flammic^is, Te.rpsiphone
perspicillata^ and Dicrurus assimilis, wei-e the bii'ds most
frequently met with. These birds were seldom seen in the
highlands.

Laniihs excubitorms seems to have a curious habit of giving a
peculiar chattering call whenevei'a wounded animal is near. We
often made u.se of this indication when ti-acking wounded beasts,
and I have no doubt of the truth of this fact.

From here I sti'uck due east through the Budonga Foi-est ngain
to the Victoria Nile. The joui-ney through this woodland counti-y
was at this time of year most arduous, all the paths being densely
overgrown with rank grass, while in the ravines the creepers and
hanging lianas were a gi'oat hindi-ance to the porters. Duiing the
four days I was in this rank jungle I saw veiy little in the way of
animal-life except Elephants, a few Baboons, and an occasional
Pviff- Adder, one of them 4 ft. 5 in. in length, and a few interesting
insects — Phasmidfe and Mantidte.

Plant-life was much moi'e interesting, and almost overwhelmiug
with its abundance of variety and its beauty.

The handsome Nightjar (Cosmetornis vexillarms) was often seen
at sunset in these forest-camps.

At length we struck the old road from Masindi to Wadelai, and
the bicycle came into use again. At my first camp along this road
there were large numbers of a golden-eyed black Weaver-bii-d
(Ploceus nigerrimus), which I saw nowhei-e else. In its size and
shape, coui'ting- and nesting-habits, it resembles very closely the
gregarious Eyphantornis cucullatus.

1903,] UIS JOURNEY TO UGANDA. 7

The Masiudi i-oad now made stiaiglit for a higli cdiiical hill,
from the slioiilder of which we had this part of Afrira laiil out
as a map l)efore us. To the south, the Budon^'a Forest; to the
west, the north end of Lake Albert, with the valley of the White
Nile extending northwai-ds ; and immediately Ijelow, from east to
west, the valley of the Victoria Nile. Descending from terrace to
terrace, we at length ari-ived at the vilLige of Fajao, just helow
the Murchison Falls, on the 22nd of August.

This wonderful gorge has been described by Baker, Yandeleur,
and othei'S, and their descrijitions are no exaggeration. One looks
down on the swirling, surging water, that, leaving the base of the
falls, sweeps round the hill on which the old fort used to stand,
with a feeling of utter amazement at the vast numbers of leaping
fishes, crocodiles, and hi])popotaiiii that have found their way
into this cul-de-sac of the Nile system.

Here 1 continued my work with moi'e success than on Lake
Albert, the counuouest fishes being Alestes baremose, A. macro-
lepldotufi, Laies ullotlcas, Clarias lazara^ Tllapia zihlii.

The natives here use enoi'uious wattle-traps, which they set in
certain fixed spots, usually out of the main force of the current,
and often catch very lai-ge fish in them.

On August 29th I stai-ted again, as this is a most unhealthy
place ; many of my men were on the sick-list, and food was getting
scarce.

Once out of the goi-ge of the Victoria Nile, we ca.me into o^ien
rolling sa-vaunali country of grass and Borassus-palms, baobabs,
and scrubby acacias. Then crossing seveiul rivers with difficulty,
we ari'ived at Wadelai on September 1st.

During this stage of the journey I noticed several birds not seen
in this pai-t of Africa before : there was Corvlnella afjinls, Paras
leucopterus, and several species of Capitonicke, all i-eminding me,
as did the laiidscape, of the Gambia on the West Coast. Here
also wei-e MeliUophaijns ballockoides, Macronyx cfoceics, Urolestes
lequatorialis, Telephonus and Craieroptus.

Fi'om Wadelai I sent my porters on to Nimule, about 100 miles
distant, taking my loads and servants down the river by boat.

After a few days' work at Wadelai, we started down the river
on the 8th of Septendier. The sceneiy on this paii of the Nile is
very charming, the hills in many places coming right down to the
water's edge. Here one sees the pi'ocess of the growth of \\.\Qsud
in every stage. Beginning with the Hoating separate plants of
Pistia stratiotes, the seeds of, first a. small floating i-ush, then of
the " 007)1 soof" grass, settle on and gradually bind together this
carpet of separate plants into a floating island of grass. 8o
a1)undant are these floating islands that often we appeared to be
stationary, even when moving at 5 or 6 miles an houi', for all the
visible lianks wei'e moving too. Once, however, the mass lodges
against the stationary papyrus, it quickly becomes overgrown by
this, and is converted into permanent sud.

Fishing villages are numerous on this pait of the Nile. The

8 MR. J. s. BUDGET! ON [Jan. 20,

natives make very good traps of papyrus-grass, and also hunt the
hippopotamus with long spears with a rope and float of ambatch-
wood attached. Amongst other fishes caught here were Mor-
myrops, Mormyrus, Hyperopisus, and Malapterxirus.

Much of this way the hills reti-eat, and thei-e is nothing seen
but grass floating and grass stationary, not even bird-life to
i-elieve the monotony. The last 20 miles, however, before reaching
the gai-i'ison town of Nimule it is very different. The Nile flows
straight towaixls the mountains above Nimvile, and here widens
into beautiful lagoons covered with water-lilies, in the foreground
sheets of Pistia of the most vivid green, in the backgrou.nd bold
wooded hills. Here and thei'e are rocky islands with schools of
hippopotami basking in the sun ; Bee-eaters {Melittophagus
jjusiU'Us), the Jacana (Parra a/ricana), and the gorgeous little
Kingfishers [Coiythoroiis cyanostygma) abound. And then the
Nile plunges into the great Nimule gorge, to tumble down
cataract after cataract, breakhig up and pulverizing the floating
vegetation, and issuing again at Fort Berkley free from sud.

I was now getting anxious about catching the Sudan Govern-
ment steamer, which comes up once a month to Gondokoro, and
determined to leave the Nile and go straight overland for
Gondokoro. The actual distance was little over 100 miles, but at
this time of year the difliculties of travelling and crossing over
rivers in flood were such that one could not tell at all how long
the journey would take. After passing through the Nimule
gorge, we came to the afliuence of the Assua with the Nile. The
Assua was now in flood, the only way of crossing being by means
of small rafts of ambatch-wood equal to taking one load at a time.
None of my porters were able to swim, and all had likewise to be
crossed on the rafts. After very neai-ly losing two men down
the rapids, the crossing was completed after eleven hours' hard
work. Here, again, the bird-life was diflferent. I saw many birds
while on the march that I was unable to identify. There were
great numbers of a Weaver-bird of brownish colour with a white
crown, building innumerable star-like nests made of straight wiry
grasses woven in at a tangent to the nest. There were also seen
in these parts, for the first time, Scojitelus notatus and Crypto-
rhina af7-a, though amongst these were not seen specimens with
red beaks as was the case on the Gambia.

Just below its confluence with the Assua River the Nile flows on
two sides of a high hill ; a fact which strikes one as remarkable,
for the two branches were mountain-torrents of very little depth
of water.

There we left the river, and passed through country with many
villages and a good deal of cultivation, especially ground-nuts and
millet. The aspect of this country of the Madis struck me as
remarkably similar to that of the Gambia : the soil was rich and
sandy, and the nuts produced were of great size. In some of the
valleys we saw quantities of very fine bamboo, while many of the
trees were almost smothered by the beautiful creeping lily
Gloriosa superha.

1903.] ms JOURNEY TO UGANDA. 9

In one of these villages quite 10 per cent, of the natixes ha<l
marked elephantiasis. They were very friendly, and provided nie
with whate\er I wanted. After three (lays' wandering hy winding
paths from village to village, we came back to the main path hy
the side of the Nile, which for over sixty miles runs along close
under a range of mountains on its western hank. From this point
northwai'ds for some time the beautifid little Pai'rot /'aknornis
docilis was common. The only Antelopes seen in this part of the
journey were Cobus leii,colis and a species of Danxdiscus. The
grass seemed to get longer and longer, and marching in the early
morning, when the heavy dew was hanging from eveiy blade of
grass in great di'ops, was most disjigrtx'able.

On September 19th we reached the flouri.shing village of a well-
known chief named Adimadi. This village was situated in a.
hollow on the top of a high hill, with natuial i-ocky foitifications
surrounding it, and overlooking a fertile valley to the east. On
the heights above the village I saw considerable nmnbers of what
appeared to be a largo red (7o/o/>?<.,s- monkey, a specimen of which
I failed to secure.

Long-horned cattle were plentiful here, and are probably the
same race as the long-horned cattle of Ankoli.

On the hill-sides were numl)ers of very fine African mahogany-
trees (yKai(i) and spi'ingsof good water. In these trees wei-e many
kinds of Plantain-eaters and Rollers (Coracias cavdaliis). This
was the first place duiing the whole journey that I met with any
Rollers. The hitherto daily rainstorms were getting less frequent,
and the diy-season was setting in.

Fiom here we marched thiough undulating pai'k-like country
with small trees, to a similar isolated group of hills, with the
village of Leju ne.stling beneath. Here, again, were fine spreading
trees, in which were numbers of beautiful glossy Starlings {S'preo
st(.perhus), and also the King of the Sparrows [Dmemelia dinemeUa).
Passing down from the Leju hills again, we marched through
country of a rather barren natiue, of i-ank grass and small
acacias. The whole way the elephant-tracks weie very numerous,
and we came suddenly on a herd of twelve with two old tuskers
among them.

The country now became more and more barren, and on
Septend)er 22nd we reached the hills again, opposite the Belgian
station of Redjaf. Here I saw several birds I had not seen before,
including Jferops niihicus, Vinarjo waalia, Laniarius erythrogaster ,
and, I think, Lanius collui-io, though it may have T)een a different
species.

At Gondokoro I sent back all my porters and LTganda servants,
and after a few days' work fishing itc, I started northwards, on
the Sudan Covernment steamer ' Abuklca,' for Khai-toum, on
Septemlier 27th.

The first few days the steamer passes through firm banks, on
which, notwithstanding the gi'ass, we saw several water-buck
and some bufialo. Many small villages line the banks, while
several old Dervish forts are passed.

10 MR. J. S. BUDGETT ON THE [Jan. 20,

At Canissa, about 100 miles noith of Gondokoro, I changed
into the ' Kaibar,' the post-boat to Khartoum ; then, passing
through the sud region in three days, we came to the mouth of
the Sobat and the land of the Shelluks.

On my arival at Khartoum, I set to work to get Arab fisher-
men and sei'vants, fishing-tackle, provisions, &c., and returned in
a few days to Fashoda.

Here I made my final attack on the PolyjJterus problem. I had
three species of Folyi^terus to work with, while material was
faii'ly abundant. However, after sevei-al weeks' work, I finally
packed up my things, and disconsolately i-eturned to England ;
having got a good deal of side-light on the life and habits of
PolyjJterus, having seen something of the Fauna and Flora of
the most wonderful river in the world, but having again failed in
my principal object — namely, to obtain the early stages in the
development of Polyj^terus.

In conclusion, I should like to say that, throughout the journey,
I received at the hands of the Uganda and Sudan officials the
most courteovis and liberal assistance on all occasions.

The following papers were read : —

1. Note on the Spiracles of Polypterus.
By J. S. BuDGETT, M.A., F.Z.S.

[Received January 19, 1903.]

On seeing a letter in 'The Field' for November 8th, 1902, by
Mr. Boulenger, in which he says that, after observing Polypterus
in captivity for more than a year, he had not been able to
learn anything concerning the use of the spiracles to Polypterus,
I determined to go over my former observations concerning these
structures and see whether I had by chance been mistaken as to
their use.

I have in captivity a pair of Polyptertis senegalus kept in an
aquarium at a temperature of 75"^ to 80° F. They are quite
tame and regularly take food from a fork.

On December 2nd I watched them for one hour after feeding.
While eating, the spiracles were repeatedly rapidly opened and
closed, though not widely. The movement was apparently inde-
pendent of other masticatory movements. Within the hotir each
of the pair came to the surface three times at irregular intervals.

1. Specimen A came to the su.rface and gulped air with the

movith ; immediately after leaving the surface, two large
bubbles of air were discharged from under the opercula.
During the descent to the bottom the two spiracles slightly
opened and from each a minute bubble of air issued.

2. Specimen B performed the same movement, but no air was

seen to issue from the spiracles.

1903.] SPIRACLES OF I'OLYPTEIIUS. 11

3. Specimen B repcitud the niovemciit, and small huliMes oF

air (lid issue from tlie spiracles.

4. Specimen A repeated tlie movement, l)iit- mi air was seen to

i.ssue from the spiracles.

5. Specimen B repeate<l the movement violently after some

e.\citi'ment, and a,s it met the siu'face loklclij opened the
spiracles, forming a triaii<fular apei'ture, one side l)ein<,'
the »side of the liead and the two other sides being the two
plates of bone whidi form the spiraculnr Hap; whether aii-
pussed in or out of the spiracles was impossible to see a«
the top of the head was out of the water. No bubbles
passed from the spiracle during descent.
(). Specimen A repeated the movement, the s[)iraclcs did not.
open, and no air was seen to issue from liiem during
descent.
On Decendjer 8th I watched them again for an hour. Puli/jituras
A and 13 came to the surface for air 8 times, and 4 times tlie
spiracles were wldiily opened above the surface of the water, and a
sound produced as of the sucking in of air.

1 have often found it convenient to kill Poli/i>l(',ran by pierinng
the crania] roof and destroying the brain. During the opi'ration
it is quite easy to stinuxlate the brain-centres in sucli a way that
tlie spii'acles ai-e widely ojiened a,s described above. It is possil)le
to stimulate continuously so that the spiracles are retained in the
widely opened condition. I believe, then, that the spiracles are
used to take in and to give out air from the swim-bladder. At
certain times tlie fish rises quite slowly to tlie surface in the
horizontal position, when it wotdd be easier for it to exdiaiige
the air in the swim-bla.dder from the surface of the head than to
turn its liead upwards in order to take air by the montli. By
closing the mouth and opercula., distending the body-wall and
ojieiiing the spiracles, 1 believe the fish is alile to inhale air, and 1
should suppose that it expires previously dui-ing the same move-
ment, as does Protopteras. I think it possible also that in the
very shallow water whicli this fish frequents at certain times
of year, it may be of use to the fisli to change the air in its
swim-bladder in this way. I have often noticed, in changing tlie
water in a tank in wliich numlieis of these fish are confined, that
wlu'ii the watei- is exhausted the spiracles are frequently opened.
The position of the spiracles almost immediately ovei' the long
slit-like glottis is in favour of the view that they are connected
in their functions witli the latter. They seem also to be used, as
I at first believed, to let out the excess of air fi-om the pharynx
after the fish has taken air into the swim-bladder, either with the
mouth or with the spiracles.

Observation upon these points is very difficult owing to the
rapidity with which the movement takes place ; but the fisli I
have been watcliing liave become very tame, after three years of
captivity, and these movements are now more slow and much
more easily watched.

12 MR. F. E. BEDDARD ON THE [Jan. 20,

2. On the Brains of Nasalis larvatns and of some other Old
World Primates. By Fkank E. Beddard, F.R.S. &c.

[Received January 17, 1903.]

(Text-figures 2-8.)

Among the types of Old World Primates not studied by Messrs.
Kiikenthal and Ziehen ^ in their othei'wise fairly exhaustive survey
of the cerebiul convolutions in the Apes, are Nasalis (if it be a
distinct genus), Colobus, and Cynopitheczi^s. I take the oppor-
tunity afforded me by the possession of brains of these thi'ee
genera to compare their charactei's with those of other Old World
genera. The memoir of Kiikenthal and Ziehen has aided me
greatly in this attempt on account of the clearness and accuracy
of their figures, many of which I have been able to verify by an
inspection of brains of the same or allied species. I do not,
however, always find myself able to agree with the selection of
characters which they u:se to define the various genera of Apes
examined by them". I cannot distinguish by any tangible
difierences the arrangement of the fui-i'ows in the genera Macacus,
Cercopithecus, Cercocehus, and perhaps Papio. It appears to me,
in fact, that among the Cercopithecidse there are only two plans
of ■ cei'ebral conformation, one confined to the Cercopithecinfe and
the other to the Semnopithecinse. The facts which I discuss in
the present communication are confirmatory of that view. But
the classificatory results to which they appear to lead are not a
little surprising, and may possibly be regarded as tending to
throw doubt upon the use of the cerebral convolutions as an
index of anything save physiological resemblances.

§. The Brain of Nasalis larvatus.

I have been able to compare the brain of the young male
Nasalis larvatus ^ with those of two other examples, which I owe
to the kindness of Dr. Charles Hose of Borneo, and I find that
thei'e are, as might be expected, some slight difierences of detail
in the arrangement of the furi'ows. The accompanying drawing
(text-fig. 2) illustrates the superior aspect of the cerebral hemi-
spheres of the specimen of Nasalis which forms the subject

1 " Untersuchungen iiber die Grossliirnfurclien der Primateii," Jen. Zeitschr. f.
Naturw. xxix. (n. s.) 1895, pp. 1-122.

2 For instance, I find that the backward prolongation of the sulcus prsecentralis,
whose absence is stated as a characteristic of Cercopithecus, is present in C. stairsi.
In Ci/nocephalus mormon the parallel fissure joins the Sylvian above, so that the
statement "a [the parallel fissure] nahert sich dem hinteren Ende von S [the
Sylvian] um S dann im Bogen zu umkreisen " is not universal in its application.
In the same species the furrow hitherto lettered b (= inferior occipital) is not
" entirely confined to the lateral convexity " ; it also extends ventrally. The fissures
u and X form a complete Y-shaped furrow as in other Old World forms. Messrs.
Kiikenthal and Ziehen write : — " x verschmilzt gar nicht oder nur scheinbar mit u."

3 Gratiolet (Mem. sur les Plis cerobraux, &c. pi. iv. figs. 1, 2) has figured a brain
of this animal.

1903.

BRAINS OF MONKEYS.

Text-fig. 2.

13

Upper suifaco otCcreliral lieiiiisplieivs of Nasalis.

a, inferior ocx'ipital fissure; c, lateral oecipital fissure; d, intraparietal :
e, postcentralis ; If, fissure of Rolando: S, Sylvian.

Text-fiff. 3.

Another brain of the same si>ccies
Letters as in text-tis. 2.

14 MR. F. E. BEDDARD ON THE [Jan. 20,

of the present communication, while the internal view of one
hemisphere is shown in another drawing (text-fig. 4, p. 15). The
variations that I have noted in the three brains are the
following : —

There is some variability in the exposure on the upper surface
of the brain of the parieto-occipital fissure. This, as will be seen
from the drawing exhibited (text- fig. 2, p. 13), is better marked in
a large female brain, given to me by Dr. Hose, than in tlie smaller
male Inain (text-fig. 3, p. 13). In the thii-d brain there are only
just traces to be observed of this fissure on the supei-ficial view of
the brain. A furrow which I identify with the postcentralis
superior is present here in rudiment (as it is in the Macaques and
othei- genera) occasionally. It is best developed, but practically
on one side only (the right), in the small male which lived in the
Society's Gardens. In the Ini-ge female l)i'ain thei-e ai-e traces
of the fissui-e on one side only (the right) ; the fissure is absent
in the third brain.

The fissure of Rolando cvirves back and joins the Sylvian
fissure in the large female bi'ain ; it does not do so in the two
remainiug lirains. On the left side of one brain the Sylvian and
the parallel fissures joined supeiioi'ly, a state of aftaii-s which is
characteristic of the brains of many monkeys.

On the right hemisphere of one brain only the lateral occipital
fissure was incomplete, and consisted of the lower arm only, there
being but a faint indication of the upper arm of this Y-shaped
fissure.

These appear to me to be the principal diflferences in the sulci
of the three brains.

I have compai'ed them carefiilly with the brains of six species
of SeTnnopithecus ', of which I owe two to the kindness of Dr. Hose,
while four were extracted from the skulls of specimens which
have lived in the Society's Gardens.

The species are as follows : — S. maurus, S. femoralis, S. hypo-
leucus, S. rubictmdus, S. priamus, and S. enielhis. The diflerences
between these brains and those described by Kiikenthal and
Ziehen ^ are slight.

The fissure lettered ^ by the above-named authors, which is
the presylvian of other writers, is less regular in its occurrence
than it appears to be in JVasalis, but the difference is not sufliciently
marked to permit of any sti-ess being laid upon the fact. In only
one instance {S. femoralis), and on one side only (the right), did
the Sylvian fissure join the parallel fissure above. The rarity
of this ai-i-angement is exactly as in Nasalis. In one case only
{S. -maurus), and also on one side only (the left), were there
indications of the fissure of Rolando joining the Sylvian fissure.
The rai'ity of this aii'nngement is again paralleled in Nascdis.

1 Kiikenthal and Ziehen (Jen. Zeitschr. 1895, p. 1) refer to the literature, but they
have not included a paper by Lankester (Quart. Journ. Sci. ii. 1865).

- " Untersuchuno'en iiber die Grosshirnf'urchen der Primaten," Jen. Zeitschr. 1895
p. 1.

1903.] BRAINS OP MONKEYS. 15

The relations of the paiieto-oecipital and tlie Simian fissures
a,p[>('a.r to me to be exactly the same in S'einiioj)ithecHS as they are
in Nasalis. As a rule, they are perfectly distinct from each
other; but in two brains the operculum was more fully developed,
as it is in the Macaques, and thus the two fissures appeared to
coincide.

The postcentralis I found to be always present, and on lioth
sides, though often asymmeti'ical.

In four Semnopithecits brains I find a sm:ill furrow running
lietween the intraparietal and the iSylvian, which may i-epresent
the anteiioi- fork of the parallel fissuie figured liy Kiikenthal
and Ziehen in several Apes (and lettered a' in their figures), and
sbited to occasionally occur in Sfimnojyithecus. The existence of
this fissure, so far as my material allows me to state, dift'crontiatos
the brain of Sonnopithecus from that of Nasalis.

Dr. Elliot Smith ' has mentioned in one brain of SeimiopitliP.cus
p.ntellasi a bifurcation of the calcai-ine fissure, such as that which
is constant, or nearly so, in Macacus and CercopiihecHS. I have
found this well developed on one side (the left) of a brain of
<S'. ruhicundus, and less m;irked on the same side of a brain of
<S'. priamiis, and on the i-ight side of a brain of S. feiaoralis. It
may be observeil, however, that in SemiioplthecH.s the T-shaped
calciii'ine fissui-e, when it is T-shaped, is not visible on a dorsal
view of the undivided brain as it is in Macacus. In the largest of
the three brains of Nasalis which I have examined, there were
indications of the same bifurcation on the light side.

Text-fig. 4.

Longitudinal median section of brain of Nusnh's.
c, calloso-marginal fissure; Ca, ealcarinc ; i.p, internal jiarieto-occipital.

The inferior occipitid sulcus is alwnys less in the Lnngui's tli.-iii
in the Miicaques. In one Semnnjntheciis ])ra,in (the left side of
;S'. prianius) this furrow showed :in unusual ch.uactei-, in th.at it
joined the parallel fissure. The s-vme occuired cm tlie right

• Cat. Pliysiol. Ser. R. C. S. ed. 2, vol. ii. p. 420.

16 MR. P. E. BEDDARD ON THE [Jan. 20,

hemisphere of a brain of S. hypoleucus. This I have not seen in
Nasalis.

This same brain (of S. priamus) and a brain of S. entellus
show on both sides an unusual condition of the inferior temporal
fissui-e, which is well developed and runs pai'allel with the Sylvian
and parallel fissures. The more usual condition in this genus is a
much shorter fissure which is more transverse in direction, and
that is the condition which obtains in Nasalis. The fissure is
not figured at all by Kiikenthal and Ziehen, but is mentioned
as being very feebly developed. Another furrow which I find to
vary in the Semnopithecits brains at my disposal is the calloso-
mai'ginal sulcus ; it sometimes bends up and cuts the surface of
the brain. In Nasalis it always does.

§. The Brain of Colobus vellerosus.

Broadly speaking the brain of this monkey is very like that
of a Macaque. There are, however, a few small points of
difference, of which one at any rate may be of some little
significance. The resemblance is so close in most particulars that
it is really unnecessary for me to describe the brain in detail.
The drawing exhibited herewith (text-fig. 5, p. 17) will adequately
prove my statement. I may, however, remark that the fissure of
Rolando quite cuts the inter-hemispheral sulcus : that the post-
centralis is well mai'ked on both sides : that the precen trails
superior is recognizable and has a direction parallel with the long
axis of the brain. On one side (the right) the Syhdan and
parallel fissures join above, as is so common on both sides with
the Macaques. The occipital lobe is very smooth, as is often the
case with the Macaques, and the lateral occipital sulcus is hardly
maiked at all. The cross-piece of the characteristically Macacine
calcarine fissure is visible when the bi-ain is inspected from above.
A small fissure, which I have not observed in othei- Old World
Monkeys, is to be seen on either side, behind the Simian fissure,
and running parallel with the longitudinal axis of the brain.
Kiikenthal and Ziehen figure, but give no name to an apparently
similar fissure (lettered E) in the bi'ain of Lagothrix humholdti.
I am able to confirm their demonstration of fact. They do not
figure the same fissure in any of the Oercopithecidse.

I am unwilling to seem to emphasize too strongly this point of
likeness between the African Colobus and the New Woi-ld
Lagothrix ; but I may remind zoologists that likenesses between
Colobus and the New World Monkeys have been pointed out.

Moi-e important than this, however, is the light which the
brain -structui'e of Colobus appears to throw upon its relationship
to other Old World genera. On account of the structure of its
stomach, and for some othei' reasons also, Colobus has been
associated with Semno2nthecus into a subfamily Semnopithecinae,
contrasted with the Cercopithecinte which embraces the remaining
Ceicopithecidee. It should be plain from the statements made in

1903.]

BRAINS OP MONKEYS.

IT

tlie present paper, and fi-om the illustrations which support them,
that so far <is its braiii is concerned Colohus cannot l)e placed
in close proximity to 6'em7W2nthecus. From this point of view,

Text-fij?. 5.

Upper surfiice of cerebral hemisplim-os of Colohus gnereza.
Ca, cilcariue fissure. Other letters as in text-fig. 2.

Text-fig. 6.

Longitudinal median section of brain of Colohus guereza.
Letters as in text-fig. 4.

Colohus is most emphatically to be placed among the Cercopithecin?e.
In fact, it is probably safer to follow those zoologists who do not
subdivide the family Cercopithecidfe at all.

Proc. Zool. See— 1903, Vol. I. No. II. 2

18 MK. F. E. BEDDARD ON THE [Jail. 20,

§. The Brain of Cynopithecus nigee.

I have been able to examine a single brain of the Celebesian Ape,
of which drawings are exhibited herewith (text-figs. 7, 8, p. 19).

The Simian fissure runs completely across and contributes to
a perfect operculum, as in the Macaques. But the brain of Cyno-
pithecus differs from the brains of the latter and agrees with that of
Semnopithecus in the fact that the infraparietal fissure of each side
bends sharply outwards before joining the Simian fissure ; this sug-
gests an exposure of the parieto-occipital, and a very faint groove
(better marked upon the left side) is probably to be looked upon
as a representative of this. The lateral occipital sulcus is a single
fissure, that is to say, it has not the Y-shape exhibited in so many
monkeys. There is, however, a faint depi-ession suggesting the
upper arm of this Y . The inf ei'ior occipital sulcus is in some respects
rather peculiar. Dr. Elliot Smith has justly pointed out ' that in
Semnopithecus ''the inferior occipital sulcus has dwindled to most
insignificant proportions, and unless the stuident examines a large
series of brains he will hardly recognize in the little arc around
the lower end of the Simian svilcus the representative of the deep
operculated infra- occipital sulcus in the Macaques." Cynopithecus
niger has an even smaller semicircular representative of this
furrow, which just arches round the lateral termination of the
Simian fissure, and is nowhere near to reaching the posterior
extremity of the occipital lobe. Further than this, a straight
furrow, either connected (left side) or nearly connected (right
side) with this, runs down the temporal lobe for some distance,
exactly parallel with and between the parallel and collateral sulci.
This I take to be the inferior temporal sulcus of brain anatomists.
In the Macaques and Cercopitheci this sulcus is (so far as my
own experience goes) quite constantly represented by a short
furrow at the lower end of the temporal lobe, and this furrow is
a marked character of those brains, and of Ce?'cocehus a,nd Cyoio-
cephcdus. This more ventrally placed furrow is not to be seen in
my specimen of Cynopithecus. The existence of an upper portion of
the inferior temporal sulcus is not, however, absolutely distinctive
of CynajntheciLS. It is indicated in a brain of Gynocephalus
mormon which I have at my disposal, and in Nasalis and Semno-
pithecus. In a brain of Cynocephalus porcarius this furrow is as
well developed as in Cyno^nthecus, but on one side only (the
right). But, with this exception, the furrow is nowhere so fully
developed as in Cynopithecxis. So far, therefore, it appears to be
characteristic of this genus. It is, however, apparently impossible
to lay much stress upon the rediiced inferior occipital sulcus of
Cynopithecus as a point of resemblance to Se-mnopithectts in the
absence of a lai'ge series of brains of the former genus. Messrs.
Kiikenthal and Ziehen figure a brain of Cynocephalus sphinx in
which the fissure is fully as reduced as it is in Cynopithecus or
Semnopithecus. I have already point-ed out (supra, p. 12 foot-

1 Cat. Phys. Series Mus. Roy. Coll. Suvg. vol. ii. p. 426.

1903.]

BRAINS OP MONKEYS.

19

note) that this state of affairs is not universal in CynocepJudus ;
but it has still got to be proved that it is universal in Cyno-
pithecus,

Text-fiff. 7. '

Lateral view of brain of Cynoplthecns nU/o:

h, inferior temporal fissure ; co, collateral ; P, parallel.
Other letters as in te.vt-ligs. 2, 3.

Text-fiir. 8.

Longitudinal median section of brain of Cj/nopithecun nigcr.
Letters as in text-fisr. 4.

The collateral sulcus, as compared with that of the Macaques
and Cercopitheci, is particularly well developed. It has, so to
speak, taken advantage of the feeble developnvent of the inferior
occipital to thrust itself forward upon the occipital lobe, where it
is phiinly visible when the brain is viewed from behind ; it is also
visible where it curves upwards upon the lateral aspect. It is
the rule among Monkeys for this fissure to be largely concealed
by the cerebellum. Its exposure for its whole length in Cyno-
pitkecus appears to me to be a characteristic feature of the brain
of that monkey.

2*

20 ON THE BRAINS OF MONKEYS. [Jan. 20,

The j9«raZZe^ sulctis extends considerably beyond the Sylvian
lis&ni'.e, and does not — as is so commonly the case with the
Macaques — join that fissure dorsally. An interesting fact about
this fissui'e is that it bends forward at its dorsal extremity, instead
of being continued on in a straight line. In this bending I see
a point <of likeness to the Semnojntheci (including Nascdis). No
absolute distinction between Cynopithectis and Macactis can be,
however, drawn on account of this fuii'ow, since Kiikenthal and
^Ziehen ^figure a brain of Macacus iiiMus in which thei'e is this
same bending forwards, and, moreover, a bifurcation of the furrow
supei'iorly, such as I note in my example of Cynojnthecits niger.
Hhejlsstm-e of Rolando presents no noteworthy characters ; it does
not iTiearly cut the inter- cerebral groove. The postcentralis is
better developed on the right side of the brain than on the left,
and is ttransverse in both cases. The 2^recentralis superior is, on
the other hand, better developed upon the left side than upon the
right. It is transverse in position. Among Monkeys this fissure
is more commonly parallel with the long axis of the brain.

The inwdian parieto-occipitcd sidcus, visible when the brain is
bisected longitudinally, presents what I regard as rather an
interesting and suggestive character. This fissui^e, in the Macaques,
&c., has a distinctly forward inclination, often making an angle
of quite 45° with the vertical. On the other hand, in the Seonno-
pitheci this furrow is nearly vertical (a brain of JVasalis) or with
a distinctly backward inclination. In Cynopithecus niger this
same furrow is nearly vertical, but with a slightly backward
inclination, thus resembling the Semnopithecidaj more than the
Cercopithecidse.

The calcarine sulctcs of Cynopithecus is not at all like that
furrow in the brains of Cynocephalus, Macactcs, Cercopithecus,
and Cercocebus. In the four last-named genera it is a T-shaped
sulcus, the cross of the T appeaidng almost, sometimes in fact
quite, upon the upper surface of the occipital lobe. This furrow is
A'ery characteiistic of those genera. In Cynopithecus the furrow
is simple and oblique in direction, as it is, as a general rule, among
the Seranopithecidee. ^

It may be convenient to tabulate the likenesses shown in the
brain of Cynojnthecus to that of Semnopithecus.

The brain I'esembles that of the Semnopitheci in : —

(1) Tlie form of the intraparietal fissure.

(2) The backward direction of the internal parieto- occipital.

(3) The absence of any junction between the Sylvian and

parallel fissures.

(4) The simple foi'm of the calcarine fissure.

(5) The shortness of the inferior occipital fissure.

(6) The presence of a well-marked superior portion, and the

absence of an inferioi- portion of the inferior temporal
fissure.

The first two charactei'S are absolutely distinctive of Cyno-

P.Z.S.1903,vol.I.Pl.I

3.

^} ;®^

3L.

1.

SsL,

^^>-

J. Green, del. et nth. . 2. lylin.terii.flros.im.p,

l.ALESTES mTERMEDIUS. 2.ALESTES OPISTHOTtENIA .
3.AMPHILIUS LONGIROSTRIS.

fin

o

>

CO
o

6^

m

N

a.

^^

5

\n
P

csi

in

%
W

H
O

w
o

V\

i:^®^"'

^^ \

P.Z.S.1903,^rol.l.Pl.IV.

//^TJ/

Itv ,

t.

V x>

/'f/'-

■<*>

^i%

"-^^

/ tl

WJ

v.-x

^-«.<'**^"

J Greeru del. etlitlL .

Mmter3a£ros .irap

MICROSYNODONTIS BATESII,

-< -h

V

M

u

<

o

m

m

o
in

o

w
ft

ffl

<:

<:

100;l] ox FISHES FROM SOUXIIEUN- CAMEROON. 21

pithecus and Senmopithecics ; the third and fourth very nearly so,
the last two characters are found also in C>/nocephalas, but are
more characteristic of Cijnopithecus and Semnopithecus.

It will be seen that if one were permitted to base a classification
upon cerebral characters, C if no pithecus would have to be renioveil
from its position among the Baboons and placed nearer to the
Langurs ; this is not too extreme an interi)retation of the braiu-
characters considered on purely morphological grounds. We may
possil)ly regard C!/iiopit]iec>i,s as occupying a somewhat basal
position with regard to Cynocephalus on the one hand, and
Semiiopllhecm on the other. For, in fact, its characters occiu-
in both, though the Uaboou-like characters are on the whole less
marked.

3. On tlie Fishes collected by Mr. G. L. Bates in Southern
Cameroon. By G. A. Boulenger, F.ll.S., V.P.Z.S.

[Ivfccivud Novemboi- 28, 1902.]

(Plates I.^V.)

The freshwittLT iish-fauna of Cameroon is still very imperfectly
known. A small list pviblished by Peters in 1876 ' and another
by Lounberg in 1895'- are the only contributions that have
hitherto appeared on this subject. Tlie collection made by Buchholz
and reported upon by Peters was important as yielding the first
specimen of the curious Fantodoii biochholzl, since rediscovered in
the Niger Delta and in the Upper Congo and Ubangi. It has now
been ascertained that this little fish flies or darts through the air,
and is, in fact, a freshwater flying-fish. .Dr. Pellegrin, of the Paris
IVluseum, has kindly informed me that, according to the notes of
M. J. de Brazza, the specimen obtained in the Coiigo by this
explorer was caught by means of a butterfly-net whilst moving
like a dragonfly above the surface of the water.

j\Ii-. Ct. L. Bates, whose previous collections included some very
remarkable Batrachians described in these Proceedings, has now
made, at my request, a rather extensive collection of freshwater
fishes in Cameroon, of which I here give a list, together with
descriptions of nine new species, one of which deserves to be made
the ty[)e of a new genus.

The specimens were obtained mostly in the Kribi River, some
15 miles from the sea; otiiers are from a small tributary of the
Campo River, near Efulen, Bulu Country, 1500-2000 feet; whilst
others again are from the Mvile River, a small straim flowing
southwards into the Campo, at about the same altitude as the
preceding.

1 Mon. Bevl. Acad. 1876, pp. 195 & 244.

- Oihers. Vetonsk.-Ak. Forh. Stockholm, 1895, p. 179.

22 MR. G. A. BOULENGER ON [Jan. 20,

MORMYRID^.

1. Petrocephalus simus Sauv.

2. ISICHTHYS HENRYI Gill. §

3. Marcusenius sphecodes Sauv.

4. Marcusenius brachyhistius Gill.

Characinid^.

5. Brycon^thiops microstoma Gthr.

6. Alestes longipinnis Gthr.

7. Alestes intermedius, sp. n. (Plate I. fig. 1.)

Depth of body 3 times in total length, length of head 4 times.
Head as long as deep, twice as long as broad ; snout shorter than
diameter of eye ; latter 2| to 2f times in length of head ; adipose
eyelid indistinct ; interorbital width ^ length of head ; width of
mouth equal to diameter of eye ; maxillary not extending quite
to below anterior border of eye ; 14 teeth (f) in the upper jaw,
8 in the outer row of the lower jaw ; length of lower border
of second suborbital less than diameter of eye, Gill-rakers
moderately elongate, 12 or 13 on lower part of anterior arch.
Dorsal II 8, above ventrals, originating a little nearer end of
snout than caudal, middle branched rays much produced (1| to 1§
as long as head) in the males. Adipose fin small, I5 to twice as
distant from the rayed dorsal as from the caudal. Anal III 19-21,
the outline very convex in the males. Pectoral shorter than head,
not reaching ventral ; latter produced into a long filament in the
males. Caudal forked. Caudal peduncle a little longer than deep.
Scales 33-34 || , 2 between lateral line and ventral. Silvery ; a
large black spot on the caudal peduncle and on the median rays
of the caudal ; latter yellow, blackish at the end ; dorsal black
and red.

Total length 85 millim.

Three specimens from the Kribi River.

Intermediate between A. longipinnis Gthr. and A. thoUoni
Pellegr, Differs from the former in the smaller scales, from the
latter in the smaller number of anal rays.

8. Alestes oPiSTHOTiENiA, sp. n. (Plate I. fig. 2.)
Depth of body 34- or 3| times in total length, length of head 4
or 4-^ times. Head a little longer than deep, twice as long as
broad ; snout a little shorter than diameter of eye ; latter 2|- or
3 times in length of head ; adipose eyelid very feebly developed ;
intei'orbital width f or nearly 5 length of head ; width of mouth
equal to diameter of eye or a little less ; maxillary not extending
quite to below anterior border of eye ; 16 teeth (|) in the upper

1903.] FISHES FROM SOUXnEUX CAMEROON-. 23

jaw, 8 in the outer row of the lower jaw ; length of lower border
of second suborbital equal to diameter of eye. Gill-rakers
moderately elongate, closely set, 20 to 23 on lower part of anterior
arch. Dorsal II 8, entirely behind base of ventrals, originating a
little nearer caudal than end of snout, twice as deep as long.
Adipose fin small, 2| or 2| times as distant from the rayed doi-sal
as from the caudal. Anal III 13-15. Pectoral as long as head,
reaching ventral or nearly so far. Caudal forked. Caudal peduncle
as long as deep. Scales 25 .t?, 2 between lateral line and ventral.
Silvery, brownish on the back ; a black humeral spot and a black
lateral band, commencing under the dorsal and extending on the
median rays of the caudal ; dorsal, caudal, and ventral fins lemon-
yellow.

Total length 130 millim.

Four specimens ; Ki-ibi Rivei- and IMvile River.

Veiy closely allieil to A.fachsii Blgr. and A. kingsleyce Gthr.
Distinguished from the foi-mev by the less massive foi'm and the
longer pectoral fin ; from the latter by the more posterior position
of tlie dorsal fin and the lai-ger eye; from botli by the more
numei'ous gill-rakers.

9. AlESTES MACROLEriDOTUS Cuv.

10. DiSTICUODUS NOTOSPILUS Gthr.

11. Xexocharax spilurus Gthr.

Cyprixiu.e.

12. Labeo axnectens, sp. n. (Plate II. fig. 1.)

Body compressed, its depth nearly equal to length of head, ih
to 5 times in total length. Head 1^- as long as broad ; snout
rounded, strongly projecting beyond the mouth, with numerous
nuptial tubercles ; eye supero-lateral, in the second half of the
head, its diameter 5 to 7 times in length of head, 2| to 3| times
in interocular width ; width of mouth, with folded lips, h to j^
length of head ; rostral fiap and posterior border of lip feebly
denticulated; inner surface of lip with numerous feeble, transverse
plica? ; a small bai'bel hidden in the folds at the sides of the
mouth. Dorsal III 9-10, with strongly notched border' ; the
longest ray equals the length of the head and twice that of
the last ; the fin equally distant from end of snout and from i-oot
of caudal. Anal 11 5 ; longest I'ay g length of head. Pectoral
rounded, a little shorter than head, not reaching ventral ; latter
I'eaching or nearly reaching vent, its first ray falling under the
ninth (sixth branched) ray of the dorsal. Caudal deeply forked,
with pointed lobes. Caudal peduncle 1)| as long as deep. Scales
36-39|^, 3 between lateral line and ventral, 12 round caudal
peduncle. Olive above, whitish beneath, with a rather indistinct
darker lateral band ; fins grevish.

Total length 210 millim.

24 MR. G. A. BOULENGEK ON [Jan. 20,

Two specimens from near Efnlen,

This new species, allied to L. parvus Blgr., is interesting as
completely connecting Laheo with Tylognathus, and showing that
the latter genus can no longer be distinguished.

13. Barbus progenys, sp. n. (Plate III. fig. 1.)

Depth of body 3| times in total leng-th, length of head
3 1 times. Snout rounded, depressed, 3| times in length of
head ; eye supero-lateral, its diameter equal to the inter-
orbital width and 5 times in length of head ; mouth large, its
width 3| times in length of head, lower jaw projecting beyond
the upper ; lips well-developed, lower interrupted on the chin ;
barbels two on each side, anterior |^ diameter of eye, posterior as
long as eye, the distance between them | diameter of eye. Dorsal
III 10, last simple ray feeble, not stronger than those following,
^ length of head ; free edge of the fin emarginate ; its distance
from the occiput less than its distance from the caudal. Anal II 5,
longest ray g length of head. Pectoral | length of head, not
reaching ventral; latter slightly posterior to origin of dorsal.
Caudal peduncle 1§ as long as deep. Scales 33^, 2| between
lateral line and ventral, 12 round caudal peduncle. Silveiy,
brownish on the back, dorsal and lateral scales brown at the base ;
dorsal and anal edged with blackish, theformer with an ill-defined
dark band across the middle ; other fins whitish.

Total length 180 millim.

A single specimen from the Kribi River.

Allied to B. hotokeri Blgr., from ISTatal. Distinguished by the
more anterior dorsal fin with 10 branched rays, the shorter anal
fin, the narrower interorbital region.

14. Barbus t^niurus, sp. n. (Plate II. fig. 2.)

Depth of body 3^ to 3^ times in total length, length of head
S| to 4 times. Snout rounded, 3| to 4 times in length of
head ; diameter of eye 4 times in length of head, interorbital
width 2^ to 2| times ; mouth inferior, its width 3 to 3| times
in length of head ; lips moderately developed, interrupted on
the chin ; barbels two on each side, anterior 1^, posterior 1|
to 2 diameters of eye, the distance between them equal to diameter
of eye. Dorsal III 8, last simple ray ossified and moderately
strong, but much thicker than the fii'st branched ray, a little
shorter than head ; free edge of the fin emarginate ; its distance
from the occiput much less than its distance from the caudal.
Anal III 5, longest ray f to | length of head. Pectoral f to |-
length of head, not reaching ventral ; lattei- a little posterior to
origin of dorsal. Caudal peduncle 1| as long as deep. Scales
24-27 ^r-, 2i between lateral line and ventral, 12 round caudal
peduncle. Olive-brown above, yellow on the sides and below ; a
series of black dots on the lateral line, and a black band on each
side of the caudal peduncle ; fins white.

1903.] FISHES FROM SOUTHERN CAMEROON. 25

Total length 120 millim.

Several specimens from the Krihi River and from Efulen,
Allied to B. camj>tacanilms Blkr. and B. potamogalis Cope.
Differs from both in the stronger third simple dorsal ray, the
more numerous scales in the lateral line (24-27 instead of 21-24),
and the coloration. From the first in the longer posterior barbel,
from the second in the longer l^rbels, the smaller eye, and the
broader interorbital region.

15. Barbus BxiTESii, sp. n. (Plate III. fig. 2.)

Depth of body equal to length of head, 3| times in total length.
Snout rounded, 3 times in length of head ; diameter of eye 5
times in length of head, interorbital width 3 times ; mouth
inferior, its width 4 times in length of head ; lips well-developed,
lower continuous; barbels two on each side, subequal, nearly
twice as long as eye, the distance between them half their length.
Doi'sal IV 8, last simple ray strong, bony, not serrated, slightly
curved, I length of head ; fi-ee edge of the fin emarginate ; its
distance from the occiput less than its distance from the caudal.
Anal III 5, longest ray | length of head. Pectoral f length of head,
not i-eaching ventral ; latter below anterior rays of doi'sal. Cai;dal
peduncle 1| as long as deep. Scales 30 ^f, 3 between lateral line
and ventral, 12 round caudal peduncle. Brownish above, the
scales darker at the base, white beneath ; fins greyish.

Total length 235 millim.

A single specimen fi-om the Kiibi River.

This is the fii'st discovei'ed West African representative of the
B. hynni group. Its nearest ally is B. tanensis Gthr., from the
Tana Pviver, East Africa.

16. Barbus kessleri Stdr.

17. Barbus guirali Thomin.

18. Barilius ubangensis Pellegr.

19. Barilius kingsley.e Blgr.

SiLURID-E.

20. Clarias liberiensis Stdr.

21. Chrysobagrus longipinnis Blgr.

22. AUCHEXOGLANIS BALLAYI SaUV.

23. Amphilius longirostris Blgr. (Plate I. fig. 3.)
Anoplopterits longirostris Bouleng. Ann. & Mag. N. H, (7) viii.

1901, p. 447.

Depth of body 9 times in total length, length of head 4 times.
Head longer than broad ; eyes small, in the second half of the
head, twodiameters apart ; interocular width | length of snout,

26 MR. G. A. boulejstger o^st [Jan. 20^

which is rounded and projects a little beyond lower jaw ; posterior
nostril midway between eye and end of snout ; pr^maxillary teeth
forming a very short band, measviring about ^ width of mouth ;
maxillary barbel | length of head, reaching root of pectoral ; mandi-
bular barbel i length of head. Dorsal I 6, nearer end of snout than
root of caudal, first ray | length of head. Adipose short, as long
as dorsal. Anal I 5, midway between root of ventral and root of
caudal. Pectoral a little longer than ventral, | length of head.
Caudal forked. Caudal peduncle 1| as long as deep. Dark olive-
brown above, mottled with black, white beneath ; dorsal, pectorals,
and ventrals light, with two transverse series of blackish spots ;
caudal whitish, with some black spots, black at the base, with a
lai'ge black blotch on each lobe.

Total length 77 millim.

A single specimen from hills in the Bulu country, near Efulen.

24. Synodontis obesus Blgr.

MiCROSYNODONTis, gen. nov.

Closely allied to Synodontis, differing only in the absence of a
free orbital border and of suborbital bones, in the more elongate
form and the rounded caudal fin, and in the curious modification
of the transverse processes of the fourth vertebi-a.

25. MiCROSTNODONTis BATESii, sp. n. (Plate ly.)

Depth of body 5 to 6 times in total length, length of head 4|
to 5 times. Body subcylindrical or feebly compressed in the
prsecaudal region, strongly compressed behind ; vent in the middle
of the total length. Head broader than deep, 1^ to 11 as long
as broad, without ridges or keels ; skin on vertex, occiput, and
nuchal shield adherent to the finely rugose bones ; a small frontal
fontanelle ; eye du-ected upwards, in middle of head, its diameter
6 or 7 times in length of head, twice in interorbital width. Lips
moderately developed ; maxillary barbel simple, nearly as long as
head ; mandibular barbels with long, slender branches ; outer
mandibular bai'bels |- length of head, inner -|. Prasmaxillaiy
teeth small, forming a villif orm band ; mandibular teeth much
shorter than the eye, 20 to 30 in number. Gill-cleft restricted to
the sides. Occipito-nuchal shield a little longer than broad,
ending in two rounded processes. Humeral process narrow,
sharply pointed, rugose. Skin smooth. Dorsal I 6 ; spine strong,
straight, striated but not serrated, about f length of head. Adi-
pose dorsal low, elongate, 2 to 3 times as long as its distance fi-om
the rayed dorsal. Anal III-IY 8-9. Pectoral spine f to | length
of head, strong and striated, with feebly serrate outer edge, with
12 to 17 strong retrorse teeth on the inner side. Ventral not
reaching anal. Caudal rounded. Dark brown or blackish, with
5 yellowish cross-bai-s above, the first on the occiput, the second
at the base of the dorsal ; the body lower down with round
yellowish spots ; throat and belly greyish, spotted or marbled wit' .

1903.] FISHES FROM SOUTHERX CAMEROON. 27

dark brown ; fins vspotted with dark brown, caudal yellowish at
the base.

Total length 100 millim.

Several specimens fi-om the Mvile River.

The vei-tebral column consists of 38 vertebrse, 13 praecaudals
and 25 caudals. The first 7 are completely united, and the
tiunsverse processes of the 4th, which form the spring-mechanism
in Synodontis, have a very extraordinary form. This process bears
a spheroidal expansion in front, whilst behind, where it presses
against the bladder, it is slightly excavated or cup-shaped. It
may be described as similar to its homologue in /Spiodontis, but
with a large, bell-shaped, bony knob attached to its anterior sui'face.
The air-bladder is large, as in Synodontis. The male genital gland
is very peculiar, being lacerated into numei'ous digitiform lobes.

Cyprixodontid.'E.

26. Haplochilus sexfasciatus Gill.

Epiplatys sexfasciatus Gill, Proc. Acad. Philad. 1862, p. 136
(Gaboon R.).

I\ccilia sexfasciata Peters, Mon. Berl. Ac. 1864, p. 396
(Liberia).

Haplochilus infrafasciatus, part., Giinth. Cat. Fish. vi. (1866),
pp. 313 & 357 (Old Calabar).

Hajilocliilus sexfasciatus Giinth. 1, c.

Lycocyprinus sexfasciatus Peters, Mon. Berl. Ac. 1868, p. 146
(Gaboon).

Ejnplatys infrafasciatus Cope, Proc. Amer. Philos. Soc. xi,
1871, p. 457.

Haplochilus infrafasciatus Steind. Notes Leyd. Mus. xvi.

1894, p. 76 (Liberia) ; Lonnberg, O^fv. Vet.-Alc. Furh. Stockh.

1895, p. 188 (Cameroon).

27. Haplochilus elegans Blgr.

OPHIOCEPHALIDiE.

28. Ophiocephalus obscurus Gthr.

Anabantid^.

29. Ax ABAS MACULATUS Thomin.

30. AXABAS PLEUROSTIGMA, sp. n. (Plate V, fig. 1.)

Closely related to A. kingsleyce Gthr., but snout longer, as long
■as the eye in the adult, at least two-thirds the interorbital
width. Dorsal XIY-XVI 10-11 ; anal VIIl-IX 10-11. Scales
27-29 f ; lateral line 14-17/10-12. A large round blackish spot
on the middle of the side, above the extremity of the pectoral fin ;
no dark spot at the base of the caudal fin.

Total length 170 millim.

Several specimens fi'om the Kribi River,

28 ON FISHES FROM SOUTHERN CAMEROON. [Jan, 20, .

CiCHLIDiE.

31. Pelmatochromis BATESII BlgT.
Recently described fi'om tlie Benito River.

32. Pelmatochromis subocellatus Gthr.

33. TiLAPIA LATA Gtlir.

Mastacembelid^.

34. Mastacembelus loennbergii Blgr.

28 dorsal spines. Length of head 3| times in its distance from
vent.

35. Mastacembelus sclateri, sp. n, (Plate Y. £g. 2.)

Depth of body 12 to 13 times in total length, length of head
(without rostral appendage) 6 to 7 times ; vent equally or neai-ly
equally distant from end of snout and base of caudal ; length of
head 2^ to 2| times in its distance from vent, and ^ to | in its
distance from first dorsal spine. Snout 3 times as long as eye,
ending in a trifid dermal appendage which is longer than eye ;
cleft of mouth extending hardly to below nostril ; a strong pi'pe-
orbital and two strong prasopercular spines. Vertical fins united
with rounded caudal. Dorsal XXYI-XXVII 85-90 ; spines
very short. Anal II 80-90. Pectoral not qviite ^ length of head.
Scales very small, 19 or 20 between origin of soft dorsal and
lateral line. Olive-brown, whitish on the belly ; a dark band on
each side of the head, passing through the eye, sometimes
continued on the anterior part of the body ; a more or less
distinct series of large, dark, light-edged ocelli along the base of
the dorsal.

Total length 225 millim.

This new species, named after our retiring Secretary, Mr. P. L.
Sclater, in recognition of many favours received from him dui-ing
his tenure of office, is based on four specimens from the Mvile
River. 3£. sclateri differs from M. marchii Sauv. in the moi'e
numerous doi'sal spines, from M. cri/jjtacanthics Gthr., liberiensis
Blgr., and loennbergi Blgr. in the larger head, and from M. con-
gicus Blgr. in the presence of only two anal spines and the still
smaller scales.

EXPLANATION OF THE PLATES.

Plate I.

Fig. 1. Alestes intermedhis, p. 22, natural size.

2. Alestes opisthot tenia, p. 22, f .

3. Avi/pliilius longirostris, p. 25, natural size.

3 a, ,, „ Upper view of head and pectoral iin, X li'.

3 6, „ „ Lower view of head, X H.

Plate II,

Fig. 1. Laheo annecfens, p. 23, f.

2. Barbus taniurus, p. 24f, natural size.

P.Z.S. J903.vo].l.P],VI.

A. •

\

\
\

\^

-'>m-'-^\

R M

/ " 7f

H

V/.K.H.del.

BaJ.e &-Dani.els3on I/" iLih-

ANATOMY" AND HISTOLOGY OF
PHAS COLO SOMA TERES .

p. Z.S. 1903, vol.1. Pl.VlI.

^A

B.T.

\\\7^ ^%

M

.L.M.

-I'M.

C. G

J c

10.

Jfi

: i

_BS

W.K.H. del.

Bale&;Dani.eJssoia,L^? lilli .

ANATOMY AN13 HISTOLOGY OF
PHASCO.LO SOMA TERES.

P.Z.S.1903,vo].I.Pl,Vin

11 .

R'M...

,**.* ^

.^.31.

--- G.C.

12

75,,?

J4.

S S

B.

s:\

/ ~-^,

1
I i

/

**\l ^
""%>

13

5 E

..SVP

:-l :l

^\

,-^y

\ \

r^^"^.

i:^^

15.

S J°

BS.

P VI

17.

u

W_KH del.

,-5

-£■

Bale & DanieisKQ-aL . litii

ANAT M.-/ AN D T i I S '1' ]:. GY OF
PHAS COLO SOMA TERES.

1903.] ON THE ANATOMY OF A NEW GEPIIYREAN WORM. 29

Plate III.

Fig-. 1. Biirhus pt'oi]enys, p. 24, ^.
2. Darhus hatesii, p. 25, ^.

Plate IV.

Microsiinndontis hatexii, p. 26.

a. The whole fish, natural size.

b. Jlouth, X 2.

c. Skeleton, natural si/e.

d. Upper view of skull, shonlder-trinlle, and imelial shieM, X Ih.

e. Side view showini; air-bladder and anterior ]iart of vertebral column, X 2.

f. Lett transverse processes of coalescent anterior vertebnu, X 3.

Plate V.

Fip:. 1. Atiahas plciirnstifl))ia, p. 27, ".

2. Mastaccmbclus sclateri, p. 28, natural size.

4. On tlio Anatomy of the Gophyrean Phai^rolosoma terfs,
n. .sp. By "W, K. Hutton, M.A., M.B., Senior Demon-
strator of Anatomy in the University of Glasgow.

[Received November 6, 1902.]
(Plates VI.-VIII.^ & Text-figure 9.)

Some tiirte ago, while engaged in ilfedging-ojierations in the
Fii'th of Clyde, Dr. J. F. Gemmill obtained two specimens of the
Gephyrean worm figured in Plate VI. fig. 1. These he lianded
over to me, and at his suggestion I undertook the task of
identifying them. My examination having led me to the coia-
clusion that I was dealing with an animal hitherto undescribed,
it seemed advisable to give some account of its anatomy ; and
this I have essayed to do in the following pages.

Unfoi-tunately the pi-esence of sand in the alimentary tract,
no less than the leatheiy natm-e of the animal's skin, formed aii
almost insuperable obstacle in the way of obtaining continuous
series of sections : as a result, my account is in some i-espects
incomplete. The worm was dredged from a depth of 60 fathoms,
nine miles to the south-west of the Corsewall Liglit. The bottom
was fine mud.

External Characters and Integuinent.

The worm is shaped like a very long-necked Florence oil -flask
and measures 40 mm. in length. Since, however, in botli indi-
viduals examined the introvert was almost completely invaginated,
in order to estimate the true length probably 8 mm. mu.st be
added to this figure. The body of the worm is seen (Plate VI.
fig. 2) to consist of three portions which difier externally, and are
found upon dissection to have definite lelations to the contained
viscera.

First, occupying the anterior third or more of the animal's

' For explanation of tlie Plates, see p. 10.

30 MR. w. K. HUTTON ON THE [Jan. 20,

length, and containing within it the more important organs, comes
fa airly stout and muscular portion 2 nun. in diameter (Plate VI.
fig. 1, A). This gradually passes posteriorly into a much slighter,
less muscular, and longer part, which extends for more than the
middle third of the length and measures 1 mm. in thickness ;
while lastly, there is a short, thick, ellipsoidal piece, measuring
5 mm. by 2"5 mm., having very thin walls which are transparent
enough to allow the coils of gut within to be dimly seen. This
terminal portion is quite sharply demarcated from the rest of the
body, and forms a sack in which the coiled portion of the intestine
(Plate YI. fig. 2, G) is contained.

The colour of the spirit-specimens is a pale yellow over the
anterior three-fourths of the animal's length ; on the thick
terminal part the tint is a dirty bluish white, the contents of the
intestine appearing through the body- wall as brownish markings.

The skin has a slightly iridescent sheen ; it is entirely devoid
of papillae and is of a leathery toughness. It is studded with the
openings of innumerable cuticular glands (Plate YII. fig. 6, C.G.).
These glands (Plate YI. fig. 3) are spheroidal, and measure '05 mm.
in diameter. Each is lined with a single layer of rather flattened
cells (L), and the majority are completely filled with small yellow,
highly refractive granules (I). The glands on the introvert are
very much smaller, and granules are not so plentiful in them as
elsewhere.

Though defiiiite papillae are not found, tegumentary appendages
are represented on the introvert by chitinous hooks. These are
ari'anged in a series of ten transverse rows (Plate YI. fig. 4),
The individual hooks vary in length (Plate YII. fig. 5) from "06
to "08 mm. ; they are brown in colour, almost black at the tip, are
but slightly curved, and the isolated hooks shown in the figure
are grooved on their concave edge for about two-thiixls of their
length (Plate YII. fig. 5, F). Each hook has a small multicellular
epidermal core. The first part of the introvert, that nearest the
" head " of the animal, is perfectly smooth, and, when invaginated,
abuts against the oviter surface of the tentacular crown through-
out two-thirds of its length ; for the remaining distance, however,
the tentacles touch the hook-bearing part. Laceration of the
tentacles during evagination and invagination is averted in some
Sipunculids by the presence of a " collar," which intervenes
between these delicate structures and the I'ough introvert. Here
there is no such Jprovision ; the end is gained hv a transverse
furrowing of the skin on the hook-bearing introvert dui'ing
invagination ; and by this means the hooks, projecting from the
base of the groove, barely come into contact with the tentacles.
An indication of this condition is represented in fig. 6, H.I.
(Plate YII.),

General Arrangement of Organs.

Figure 2 (Plate YI.) is a drawing of a dissection of the worm.
It has been opened along the ventral aspect, and the two retractor

1903.] ANATOMY OP A NEW GEPHYREAN WORM. 31

muscles (D.R., V.R.) of the left side have been cut through. The
anterior part of the alimentary canal has been twisted aside and
turned so that it presents to view its dorsal surface. The intro-
vert is almost completely invaginated, the real "head" of the
worm being about the position of the two eye-spots (E).

The internal aspect of the body-wall has a glancing appearance
like mother-of-peai'l ; the longitudinal muscles form a continuous
sheet with no di\-ision into strands. Thei-e are four i-etractor
muscles (D.R., V.R.), two dorsal and two ventral. The latter are
more than twice the length of the former, stretching fi'om the
termination of the anteiior muscular pai-t of the body (between
A & B, fig. 1) to the anterior pait of the cesophagns, where they
fuse with the doi'sal miiscles to form a sheath for that tube. They
arise, one on either side of the nerve-cord. The dorsal muscles
are attached to the body-wall on either side of the hind-gut, a
little way behind the anus. In fig. 2 the only part of the
nervous system seen is the eye-spots, which are visible in a
dissection as a couple of black specks shining througli the walls
of the inti'overt.

The alimentary canal (G.F.), with the exception of the anterior
part of the a?sophagus, is shown in full. It is, like the worm
itself, divisible into three parts, an anterior and a posterior
portion, comparatively straight, and a middle jiiece (G) exceed-
ingly coiled.

In the drawing, for the sake of clearness, that part of the gut
(F) contained in the thin portion of the worm is shown as though
absolutely straight ; it is really slightly crumpled. Nowhere is
there any intestinal spiral ; thei'e is no spindle-muscle, and the
intestine is entirely free except for about 2 mm. behind the anus,
where the i-ectum is attached to the body-wall by thin radially-
disposed strands of muscular fibi^e, which in sections (Plate VII.
fig. 6, M) appear to divide the crclom into dorsal and ventral
portions. The nephridia are noticeable objects in a dissection
(fig. 2, B.T.) on either side of the anus. All that is visible of
the blood-vascular system is the tortuous dorsal vessel (H). The
generative organs are visible only in mici'oscopic prepai'ations.

Ihiscidahire.

The dermo-muscular tube has the arrangement common to the
Phascolosomidaj. Underneath the integmuent there lies a layer
of circularly-disposed fibres (Plate VII. fig. 6, R.M.), and beneath
this again one of fibres having a longitudinal dii'ection (L.M.).
As already mentioned, this lattei' layer is a continuous one, witli no
division into strands such as are seen, for example, in the genus
Sipunculas. The musculature is thickest and sti'ongest in the
anterior third of the animal's body ; becoming thinner as the
body narrows, it reaches the extreme of tenuity over the ellipsoidal
posterior part, the thinning of tlie tube appealing to take place at
the expense of the circular layer.

32 MR. W. K. HUTTON ON THE [Jan. 20,

Around tlie anus and the external nephridial apertures, both
layers furnish special sphincter-fibres ; while along either side of
the nerve-cord, as is afterwards noted, there runs a slender longi-
tudinal strand of muscle-fibre (N.C., fig. 6).

The genei'al arrangement of the four special retractor muscles
of the introvert and of the radial rectal muscular attachments
have already been described ; there remains only, in regard to the
former, to speak of the special appearance presented by the
muscular fibres in this worm. On comparing the ventral
retractors with the general longitudinal musculature, one is
struck by the great similarity of texture presented by both. A
glancing, bright suggestion of firmness, I'eminding one of the
surface of fresh tendon among the higher forms, is noticeable.
The doi'sal retractors, on the other hand, have a dull, almost a
translucent appearance. Microscopical examination of transverse
sections reveals little difierence between the two sets of muscles,
save only that in a given area fewer muscular fibres are present
in the dorsal muscles, and the amount of intei'muscular connective
tissue appears to be greater than in similar sections of the ventral
set (Plate VII. fig. 7, A & B). On the whole, both from the
relatively marked shortness of the dorsal muscles and from the
appearances just described, I incline to suspect that they cannot
be functionally veiy active. They contain a sufiicient amount of
connective-tissue between the proper muscular bundles to make
the suggestion that they are in a state of metamorphosis and
regression at least probable {vide J. Bland Sutton's ' Ligaments,'
&c., 2nd ed., 1897, p. 3 et seq.).

Alimentary Caned.

Surrounding the mouth, which lies at the bottom of the depres-
sion formed by the union of their bases, are several simple
leaf-shaped tentacles. So far as could be ascertained by the
examination of a series of sections, 1 5 tentacles are present, and
of these 11 are longer than the others. Each tentacle contains a
branch of the blood-sinus and a slender nerve (Plate VII. fig. 8).
This figure, which represents a typical transverse section of a
tentacle, shows that on the mesial aspect the cells covering; its
surface are nari'ow, long, arranged in palisade form, and are
ciliated, while those of the opposite side form a simple columnar
epithelium. For the gi-eater part of their extent the tentacles
ai-e free, but round the mouth their bases ai'e fused together, and
form a series of gutters which lead to the oral apeiture.

The mouth itself lies on the top of a papilla, whose sides are
grooved by the above-mentioned furrows. The cells lining the
o-rooves are cubical and ciliated, and are continuous with the
ciliated epithelium on the mesial aspect of the tentacles. Succeed-
ing the mouth is a short oesophagus with veiy muscular walls ; in
section (Plate VII. fig. 10) it appears to be star-shaped, the lumen
of the tube being encroached upon by the inward projection of a

1903.] AXATOMY OF A NEW OEPHYREAN AYOUM. 33

series of papillfe whose epithelial covering is apparently devoid of
cilia.

For the reason mentioned at the beginning of this paper, it was
no easy niattei- to gain a cei-tain knowledge of the histology of the
alimentary tiact. Figure 9 A (Plate VII.) shows the character of
the epithelium of the mid-gut geneiully : lying upon the thin mus-
cular wall are nai'row cells of varying length, with a basal nucleus
and granular c}i;oplasm ; the outlines between the several cells
were in many j^laces most indistinct. Tlie epitheliiun of the rectum
(fig. 9 B) is composed of cul)ical cells, witli a distinct margin on
the surface next the lumen of the gut. They possess cilia,, and
were the only cells in the alimentary ti'aet in which the pr&sence
of a ciliated boi'der could with certainty be determined.

Circalatonj Si/stem.

This consists of a doisal vessel, a sinus sui'rouniling the base of
the tentacles, and tentacular vessels.

The dorsal vessel is single, and though convoluteil it is devoid
of diverticula : coursing along the dorsjil wall of the fore gut, it
surrounds the base of the tentacles with a cii'cular sinus into
which the brain dips (Plate VIII. figs. 11, 15, IG) ; from tliis
circular vessel tentacular sinuses arise, they form noticeable
objects in sections of the tentacles (Plate VII. fig. 8, E.8.).

Nervous System.

This is of the usual type, and consists of a biuin, a ventral
nerve-chain, and a couple of sense-organs.

The bi-ain is a small, somewhat cordiform body, measuring in its
long axis, which corresponds with that of the animal, three-tenths
of a millimeti'e ; it lies on the dorsal wall of the pharynx at the
base of the tentacles ; it is suri'ounded by a fine capsule of con-
nective-tissue, and the anterior surface in addition is covered by
ail epithelium (Plate VIII. fig. 13, E.), which is continuous with
that of the outer non-ciliated aspect of the tentacles ; laterally, the
muscular tube formed l)y the union of the reti'actoivs abuts u})on it
(fig. 15, R.M.); below the brain is the blood-sinus (B.S., figs. 11,
l8, 15) from which the tentacular vessels arise.

The majority of the cerebial cells are small, with a clear cyto-
plasm and i-elatively large nuclei ; they are massed anteriorlj',
dorsally, and postei'iorly : the central and greater part of the
ventral aspect of the brain are composed of fibres, l)ut behind
the point of origin of the circumresopliageal commissures there
ai-e to be seen in the lowei- coitical pai't sevei'al very lai'ge
ganglion-cells (G.O., fig. 11): these giant cells measure "03 mm.
in diameter; some are pyriform, others iri'egular in shape, and
these appear to be multipolai-, unlike those of Fhyraosoma
(Shi])ley, No. 7.^), but on this point I cannot 1)6 absolutely
definite.

' This iiumlicr luis vcfijn'iicc to flio list i>f iiutliovs quoted on p. '10.

Piioc. ZooL. Sue. -1903, Vol. I. Ko. III. 3

34 MR. W. K. HUTTON ON THE [Jail. 20,

Fi'om the brain two pairs of nerves arise. The fii'st pair
(Plate VIII. fig. 16, B), which are vei-y slender, take their origin
from the ventral asj)ect of the ganglionic mass jvist anterior to the
eye-spots ; from them twigs pass to supply the epineural canal and
apparently also the two dorsal tentacles. The remaining tentacles
receive their nerves from the two stout cii-cumoesophageal con-
nectives (fig. 16, A), which constitute the second pair of nerves,
and arise ventro-laterally behind the eye-spots, coui'sing round
the base of the tentacular crown, between it and the muscular
tube formed by the fusion of the retractor muscles. Yentrally,
the cii'cumoesophageal commissures unite to form the nerve-cord.
This shows no trace of segmentation. Dorsally it is composed of
fibres, while ventrally nerve- cells are distributed evenly along its
whole length. It is accompanied laterally by two fine muscles, as
is usual in the Sipunculidte.

Sense-Organs.

In tracing back series of sections, there becomes apparent on
the dorsal wall of the pharynx a patch of long-celled columnar pig-
mented epithelium. The cells of this layer are found (Plate VIII.
figs. 12, 13, 14, 15, S.E.) to be the dorsal wall of a slit- like canal,
whose ventral wall is formed by cells which, on the one hand, cover
the brain in fi-ont, and on the other ai'e continuous with the non-
ciliated epithelium which clothes the outer aspect of the two
dorsal tentacles (S.V.), which are seen in fig. 12 to have united
along their mesial margins.

Posteriorly the canal thus formed (which may be called the
epineural canal) becomes very narrow centrally, and ends blindly
by dividing into two little culs-de-sac (S.P., fig. 15), Avhich
project downwards and outwards into the cerebiul cortex, and are
lined by a columnar epithelium the cells of which are deeply laden
with granules of a reddish- black pigment.

The cup-like organs thus formed are usually called " eye-spots,"
and are present in at least four other species of Phascolosoma, but,
so far as I am aware, no account of their structure has hitherto
been published. Their existence in an animal which lives at a
depth so great as that of 60 fathoms is noteworthy, and appears
to me to ai'gue for the organs in question some function other than
even merely photometric ; but a small fraction of difi'used daylight
can permeate so far. They are always to be found in the genus
Phymosoma^ and from the above-given description it will be seen
that the sense-organs of Ph. teres resemble in the main those of
Phymosoma varians (Shipley, 7). They differ, however, from
those of Phymosom-a in their relations to the tentacles.

In the genus Phymosoma the cells entering into the formation
of the eye-spots are continuous, on the one hand, with the epithelium
covering the anterior aspect of the brain, and, on the other, with
those which constitute the lining of the mesial surface of the
tentacles. In Phascolosoma, on the contrary, the sensory epi-

1903.] AXATOMY OF A XEW GEPHYREAN WORM. 35

thelium is furnished by a tentacular sui-face whicli is apparently
external.

This seeming discrepancy is readily explained by a reference to
the following diagrams (text-fig. 9), modified from Selenka, in
which A represents the position of the mouth, B that of the
cerebral ganglion, while thick-dotted lines indicate the relations
of the tentacular wreath.

In text-fig. 9 a this wi-eath is seen surrounding the mouth with
a simple circlet, as in Phascolosoma teres. In such a case the
infundibuliform approach to the cerebral organ must be lined by
cells, continuous not only with cells of the introvert epiilermis
but with those covering the outer aspect of the tentacular
chaplet.

B

Text-fig. 9.

y»

\ -s

' \ / I B I \

/ \

I
i
/ \ A

I A

\

\

\ /

a ^

Tentacular wri-ath of {a) FMxcolosoma and (i) FJii/mosoma.

In text-fig. 9 b the crown of tentacles is seen to have been in-
dented in its upper (neural) part, and the edges of the indentation
have been bent up in such a manner as to form a hoi-seshoc-
shaped lophophore which, partially encircling the brain, overhangs
the mouth instead of surrounding it. The finely dotted line in the
diagram represents the position of the "lower lip" in Phi/mosoma ;
this may be looked upon as corresponding to the fused bases of
aborted tentacles. Here, then, the brain, lying at the base of the
lophophore, can only be approached by a canal whose walls are
coated by an epithelium continuous with that covering the mesial
suifacc of the tentacles.

The arrangement seen in Phctscolosoma is the more primitive.
Briefly stated, tentacular surfaces which are inner and outer in
the genus Phascolosoma, become outer and inner respectively in
Phymosoma, owing to the secondary indentation of the crown of
tentacles.

According to Shipley {loc. cit.), the space included within the
concavity of the lophophore of Phymosoma becomes the repre-
sentative of the " praM)ral lol)e " of Phoronis.

:^36 MR. W. K. nUTTON ON THE [Jan. 20,

Altliougli nowadays, in virtue of our knowledge of the totally
distinct modes of development of the tentacles in Phoronis and
the Sipunculids, it does not seem possible to institute any
homology between the lophophoral crowns of Phoronis and
Phymosoma, yet it may be of interest to note that in the species
of Sipunculid under consideration the homologue of the pigmented
epithelium, which in Phymosoma covers Shipley's " prfeoral lobe,"
is to be found in that layer of pigmented cells (Plate VIII.
fig. 12, S.E.) which forms the dorsal wall of the epineui-al canal
as well as in that layer continuous with it, bounding the anterior
svirface of the brain (fig. 13, E.).

Nejjliridia.

These are small (Plate A^I. fig. 2, B.T.), measuring less than the
twentieth part of the length of the woim ; the right nephridium
is the longer and more convoluted of the two. Each organ is
retort-shaped (Plate YIII. fig. 17), having a bulbous portion (0)
attached to the body-wall, and a tubular part (D) which, arising
from the postei'ior part of the bulb, bends i-ound it mesially and
pi'ojects freely forwards into the coelom. The wall of the bulb is
thin, covered externally by the cells of the peritoneum (A, fig. 18),
and intei-nally by a single layer of granular-looking columnar
epithelium (0, fig. 18); it is composed of slender intercrossing
muscular bands. The cavity of the bulb is packed with fine
vesicles (E, fig. 18), which, it will be seen, are derived from the
secreting epithelium of the tubular portion of the nephridium.
The bvilb communicates with the exteiior by a fine canal (A, fig. 17),
lined with a small-celled epithelium, which pierces the body- wall
immediately in front of the anus, and its cavity is, further,
placed in connection with the coelomic space through a slit-like
opening (B, fig. 17) in front of the excretory canal. The lips of
this opening, lined with ciliated epithelium, lie between the body-
wall and the bulbous part of the nephridium, and are concealed
by its overhanging anterior extremity. The tul^ular portion of
the organ has a wall which is formed by a peritoneal lamella,
beneath which is a layer of muscles (B, fig. 18). This muscular
layer consists of a meshwork of strands, between interstices of which
the lining-epithelium bulges outwards, forming little crypts lined
with a single layei- of secretory cells (0). In cross section there
are seen projecting into the lumen of the tube feathery columns
of cells (D, fig. 18), which have an appearance comparable to that
of goblet-cells. The nucleus is small and irregular in shape,
placed basally ; while the cell-contents may be seen to become
clearer near the free margin, and a.i-e evidently extruded ultimately
as the thin-walled vesicles which are found filling the lumen of
the organ.

Shipley aptly compares the process, which takes place also, in
Phymosoma, to the formation of the granules exci'eted by the
cells of the mammaiy gland.

1903.] ANATOMY OF A NEW OEPnYREAX WOItM. 37

The Generative Oryaus.

Sti-etcliing across the ventral aspect of the cfeloin, at the point
where tlie ventral retractors are attached to the hody-wall, is a
Land of cells with relatively large nuclei. These cells const itule
the genital hand and ai-e derived from the peritoneal epitheliuiu.
In the single worm examined, neither free o\'a nor spermatozoa
were to be seen in the body-cavity.

Systematic.

The presence of hooks and the possession of four i-etractor
muscles are characters which at once place rJtascolosoina teres in
the first of the five " groups" into wliich Selenka, (5) divides the
genus. At the ti)ue of publication of his ' Die Sipuuculiden ' three
species alone [Ph. virjgare, Ph. elongatum, and Ph. cjjlindratmn)
composed the gi'oup. To these must now be added, as well as
J'h. teres, Ph. saadcri Collin (1) and Ph. lobostomum Grube
(Fi'-clier, 2). Tlie position of Pltascolosoma ahnoriais Sluiter (ti)
must in tlie existing state of classification remain doubtful, as the
whole antei'ior end of the animal was wanting.

The neai-est ally of Ph. teres in the group is Ph. elongatum
(see Kefei-stein, 3). Both have the same arrangement of tentacles
and the same number of rows of hooks, while the hooks themselves
are very similar. Ph. teres, however, difl^'ers widely in the pro-
portionate length of its introvert, and, as regards its interuiil
anatomy, in the small size and unequal bulk of the nepluidia,, in
the absence of any definite spiral coiling of the intestine, in the
great length of the hind-gut, and in the absence of a spindle-
muscle ; and, as well as in these particulars, in its characteristic
habit of body. Fiom eveiy other member of the group except
7Vi. ci/lindrat/i')ii it (lifters in the complete absence of papilhe.
These peculiarities warrant the systematic zoologist in assigning
to the animal under discussion a distinct pLice in the genus, a.nd
I propose for it the name of Phascolosoma teres.

That the genus Phascolosoma is an eminently variable one is
evident to anj^ investigntor who makes a systematic study of tlie
Sipunculidai ; and that this tendency to variation may result in
the inclusion in one small sub-group of animals which difter in an
ast(mishing extent in external habit, is abundantly exemplilied
by the fact that Phascolosoma teres and Ph. sanderi (Collin) — two
Avoi-ms so dissimilar that a. casual observer would be pai'doned for
placing them in difterent genera — must yet, in the present rather
luisaiisfactory state of classification, be united in the group cnlled
by Selenka ''(iroup 1." of the Phascolosomidic ; yt^t, while tln^
latter was considered by its describer Collin to pi-esent points of
i-escud)lance to that curious 8ipunculid Goljingia niacintoshii (8),
;Mid seems to have points in common with A.y)idosiphon, the
foniiiT appears to me to }»resent in its structui'al pecidiarities
definite alEnities to the sedentary Phasculioii. lloule (4), who

38 MR. W. K. HUTTON ON THE [Jail. 20,

investigated the Phascolosomiclse obtained by the ' Ti^availleur '
and ' Talisman,' in speaking of classification, goes the length of
saying, " L'espece bien affirmee n'existe pas dansle genre Phasco-
losome " ; and by this somewhat strong assertion he would
indicate that the genus is in such a plastic condition— in a state
of fluctuating variability as one may term it — that it is capable
of being influenced by what Semper called " the conditions of
existence " in such a way, that within the genus we may have
numbers of types, differing on the whole but little from one
another, the links of their genetic affinity easily supplied, which
yet point the way toward the probable origin of the other
Sipunculid genera. As Roule expressed it : " Oe genre est encore
en voie de transformation, il se morcelle en un grand nombre de
types pevi dissemblables, et ne se divise point en especes precises,
dont les variations s'enserrent dans des limites peu eloignees."
Roule's investigations led him to the conclusion that the Phasco-
losomidas examined by him formed a series of links connecting
the parent Phascolosoma stock with Aspidosiphon. In the phylo-
genetic table drawn up pi'ovisionally by Selenka in his beautiful
monograph ' Die Sipunculiden ' (page 6), a polyphyletic origin is
suggested for both Phascolion and Aspidosipho?i. The former
may have arisen dii'ectly from the parent Phascolosomid, or from
a common stem (similarly derived) which branches into Onchne-
soma, Tylosoma, and Phascolion ; while Asjndosiphon may have
been derived from the directly produced Phascolion, or from a stem
common to itself, Phyniosoma, and Cloeosijyhon.

In Phascolosoma teres we have, as it appears to me, not perhaps
an actually intermediate type, but at the least a species which
gives us hints as to the probable path taken in its phylogenetic
course by Phascolion ; while the species examined by Roule play
the same role as. regards the genus Aspidosiphon. Phascolosoma
sanderi, though considered by Collin to approach Golfingia, has a
distinct but superficial I'esemblance to Aspidosiphon, and was
rightly, as I think, placed by its describer in the genus Phascolo-
soma.

In the worm which forms the subject of this paper the follow-
ing characters suggest Phascolion-like affinities : — The differing
bulk of the two nephridia, the relatively feeble development of
the dorsal retractors (a character upon which Roule placed much
reliance in his investigations), the lack of spindle-muscle, and the
coiled, but not spirally-wound gut, — all these are peculiarities
which place the worm in closer relationship with Phascolion than
• with any other of the groups derived from the original Phasco-
losomid stock.

Information as to the habits of Phascolosoma teres is wanting ;
but in its mode of life it is probably sluggish, lying buried in soft
ooze with its anterior end alone projecting. The thin- walled
bag-like posterior portion, endowed with but feeble muscularity,
so sharply demarcated from the rest of the body, so relatively
large, would undoubtedly act as a hindrance to active locomotion

1903.] AXATOMY OF A XEW GEPHYREAN* WORjr. 39

througli sand oi^ mud. Most species of Phascolton live with tlio
posterior part of tlie body either einlieiUled in mud or protected
by some Ci\st-off mollusk-sl\ell ; and if we assume, as the external
habit of Phascolosomn teres seems to me to warrant, this similarity
of mode of life between the two typos, certain peculiarities of
structure in Ph. teres become of easy explanation.

Selenka, long ago, pointed out that those Sipuncnlids which
live in shells usually have only one nephridium. In other wt)rds,
in animals of sluggish habit, the needs of metabolism are served
by one excretory tube.

Now in Ph. teres, while there are two nophridin, they nre both
small and ai'e of difi'ei-ent sizes ; the left has I'eti'ograded to a
greater extent than the right. It is to be noted thsit when only
one nephi'idium persists — as in PhascoUoa — it is that upon the
I'ight side.

Enfeeblement of the muscular system is correlated with inactive
habits ; and in worms lying half buried in mini a. strongly
developed system of retractors censes to be a clamant need. The
extreme case is reached in Phascolion, where often the ventral
I'eti'actoi's alone remain, while an indication of regression of the
doi'sal i/eti'actor system is seen {vide supra) in Phascolosoma teres.

Lastly, in this worm, the coils of intestine, completely filling
np the sharply-delimited posterior part of the body, could scarcely
be efficiently acted upon by any spindle-muscle, which, we see, is
absent.

In these structund chnracters, then, we seem to oljtain in
Ph. teres an indication of the path taken by Phascolion in its
derivation from the more primitive Phascolosoma.

Summary. — With the characters of the Group,

Phascolosoma teres, n. sp.

Body 40 mm. long, introvert ca. 8 mm. Posterior portion
stout, sharply demarcated, ellipsoidal, containing the coiled
portion of the gut.

Colour yellowLsh, merging to dirty bluisli white over the
postei'ior pait.

Skin smooth, iiidescent, devoid of papUlfe, translucent posteriorly,
beset with numberless glands. A simple wreath of 15 tentacles
surrounds the mouth. Forward on the introvert 10 rows of
backwardly directed, slightly curved, brownish hooks, •()6--08 nnu.
long.

Two eye-spots. Alimentary canal free, without spindle-muscle
or mesenterial attachment save close to the anus. 8[)iral winding
absent, but the tulie much coiled in its middle portion. Hind-
gut very long. Two nephridia, the right the larger, one-twentieth
of the body-length. Simple contractile vessel.

Hab. Firth of Clyde. Dredged from 60 fiithoms.

40 on the anatomy of a new gephyrean worm. [jan. 20,

Works referred to.

(1) A. Collin. — " Gephyreen gesammelt von Herrn Stabsarzt

Dr. Sander," in Archiv fiir Naturgeschichte, Iviii. p. 197,
1892.

(2) Fischer. — " Die Gephyreen des Naturhistorisclien Museums

zu Hamburg," in Abhandlungen aus deni Gebiete der
Naturwissenschaften &c., Hamburg, Bd. xiii. p. 1, 1895.

(3) Keferstein.— " Beitrage zur Kenntniss der Gattung Phascolo-

soma F. S. Leuck.," in Zeitschrift fiir wiss. Zoologie,
Bd. xii. p. 35, 1862.
(3 a) Keferstein. — " Beitrage zur anatomiscben und syste-
matischen Kenntniss der Sipunculiden," in Zeitschrift fiir
wiss, Zoologie, Bd. xv. p. 404, 1865.

(4) RouLE. — " Sur le Gephyriens des grands fonds de la mer

recueillis par le ' Travailleur ' et le ' Talisman,' " in Oomptes
Rendus, cxxvii. p. 197.

(5) Selenka. — " Die Sipunculiden," in Semper's ' Reisen.' Wies-

baden, 1883.

(6) 0. Ph. Sluiter. — " Beitrage zu der Kenntnis der Gephyreen

aus dem Malayischen Archipel," in Natuurkundig Tijdschrift
veor Nederlandsch-Indie, Deel xlv. p. 472, 1886.

(7) Shipley. — " On Phymosoma varians" in Quart. Journ. Micr.

Sci. vol. xxxi., n. s. p. 1, 1890.

(8) Lankester — " Golfingia macintoshii, &c." Trans. Linn. Soc.

Lond. 2nd ser., Zool. vol. ii. p. 469, 1885.

EXPLANATION OF THE PLATES.
Plate VI.

Fig. 1. The entire animal, magnified about 6^ times. A, the anterior muscular part.

B, the middle thinner portion. C, the bag-like posterior piece. AN., the
anus. pp. 29, 30.

Fig. 2. Semi-diagrammatic figure of a dissection, magnified about 13 diam. I, the
introvert. E, the e3'e-spots. B.T., the uephridia. H, dorsal vessel.
D.R., dorsal retractor muscles. V.R., ventral retractors. AN., position of
anus. Gr, coiled part of gut. F, straight part. pp. 30, 31.

Fig. 3. Cuticular glands in vertical section. X 290. E.M., radial muscles.

C, cuticle. L, secretory cells. T, granular contents, p. 30.
Fig. 4. Hooks from introvert. Surface view. X 154, p. 30.

Plate VII.

Fig. 5. Isolated hooks. X 290. F, furrow, p. 30.

Fig. 6. Transverse section 1 mm. behind the anus. X 53. C.G., cuticular glands.
R.M., radial musculature. L.M., longitudinal do. M, muscles attached
to rectum. Gr, rectum. B.T., nephridia. T, tentacle cut across.
H, hooks. I, introvert. N.C., ventral nerve-cord and muscles, p. 31.

Fig. 7. Histology of the retractor muscles (Powell & Lealand yV-imm., eyepiece A).

A. Transverse section of dorsal retractor, showing amount of interstitial

connective between the bundles.

B. Ditto of ventral retractor, showing almost no intermuscular connective.

p. 32.
Fig. 8. Transverse section of tentacle. X 290. B.S., blood-sinus. N, nerve.

A, ciliated mesial cells. B, columnar outer cells, p. 32.
Fig. 9. A. Epithelium of mid-gut. X 290. B. Epithelium of hind-gut. X 290.

p. 33.
Pig. 10. Transverse section of oesophagus. X 290. Part only of the stout muscular

wall is shown. M, muscular wall. E, the epithelium thrown into papillae.

p. 32.

190'].] ox A XEW FRESnWATEn CRAB FROM Uri'EK GTIXEA. 41

Plate VIII.

Fijj. 11. Transverse section of brain behind origin of ccsoiiliageal cotnTnissures, to
show triant ganglion-cells. X 290. K.M., retractors. 13.S., blood-sinus.
G.C., giant cells, p. 33.
Figs. 12, 13, 14, 15. Four transverse sections through the brain and base of the
tentacular crown, to show the epineural canal and formation of the cyc-
spots. X 53. B.S., blood-sinus. V.C., nerve-cord. K.M., retractor muscles.
S.E., sensory epithelium of the dorsal, and S.V., that of the ventral wall of
the epineural canal. T, tentacles. S.P., eye-spots. E, epithelium of
anterior cerebral surface. ]). 31.
Fig. 16. Transverse section through brain and bases of the tentacles, showing the
origin of the circuuuesoiihagcal commissures (X 53 and reduced), p. 31.
A. Circunuesophageal nerves.

15. Nerves to the epineural canal and dorsal pair of tentacles.
C. Two nerve-strands, in cross section, which seem to be connected with

the sense-organs.
T). Fused retractor muscles.
K. IJlood-sinus.
Fig. 17. Diagram of a nephridium. A, excretory canal. B, coclomic pore.

C, vesicle. D, tubular part. E, body-wall. p. 36.
Fig. 18. To show histology of tubular portion. X 290. A, peritoneal cells.
15, muscular wall. C, secreting cells which at D are seen forming
feathery columns. E, vesicles, p. 36.

5. On Potamon (Fotamonaufes) latidactylani, a new Fresh-
water (,'ral) from Upper Guinea. By Dr. J. G. de Man,
of lerseke, Holland.-^

[Received November 15, 1902.]
(Plate IX.'O

In the year 1881, Poto.?»o?? o/rica7m?>i A. M.-E. was known only
by the short diagnosis and the figures in the ' Nouvelles Archives
du Museum,' ' made from a quite young individual from the
Gaboon. The anterior legs had neither been described nor figured,
and it is therefore not surprising that some older specimens of a
Fotamoit from Liberia were referred l)y me erroneously to this
species \

Some time since, three adult specimens of a Potamon were sent
me for examination by Prof. Jefirey Bell ; they had been collected
in the River Prah, in the south of Ashanti, West Africa. These
Crabs not only proved to belong to the same species as that
described by me in 1881, when compared with a female of medium
size and a veiy young male from Libei-ia in the Leyden Museum,
but they proved also to be new, as a typical specimen of P. afri-
caimm, a middle-sized female from " Ogoue " (evidently the
River Ogowe, just below the Equator), was khidly sent me by
Prof. Bouvier, and as a more complete description of P. africanum
was published in 1887, in which, however, the legs have not licun

1 Communicated by F. Jeffeey Bell, F.Z.8.

- For explanation of the Plate, see p. 4-7.

3 Vol. V. p. 186, pi. xi. fig. 2 (1869).

* Telpkusa africana de Man, Notes Leyd. Mas. iii. p. 121 (1881).

42 DR. J. G. DE MAN ON A NEW [Jan. 20,

figured (A. Milne-Edwards, ' Observations stir les Crabes des
eaux douces de TAfrique/ Paris, 1887, p. 4, pi. 2. fig. 8). I
therefore propose the name of Potamon latidactylimi for this new
species that inhabits Liberia and Ashanti, on account of the
characteristic shape of the hands, and as this feature was not alluded
to in my somewhat incomplete description of 1881, another will,
I think, be welcome.

The rivers of West Africa and of the Soudan are inhabited by
several species of PotaTiion which differ from the other species
of this subgenus in the existence of tioo epibranchial teeth behind
the external orbital angle, instead of one as is usual. I was
at first inclined to create for these species a new subgenus, but
the difi'erences from Parathelphusa are, indeed, of too little
importance. These species are the following : —

Potamon auhryi H. M.-E. Gaboon.

,, peJii Herklots. Gold Coast.

,, africanum A. M.-E. Gaboon ; French Congo.

,, decazei A. M.-E. French Congo.

,, emarginatibm Kingsley. West Africa.

,, floweri de M. Soudan.

,, latidactyltim, n. sp. Tipper Guinea.

It must, however, be observed that P. emarginatum is con-
sidered by Miss Rathbun (Proc. U. S. National Museum, xxii.
1900, p. 283) to be the same as P. auhryi. P. latidactylum
may at first sight be distinguished from P. africanuvi hy the
different shape of the extraorhital and epibranchial teeth and hy
the different form of the hands, especially of the fingers, but there
are still other difi'erences.

The cephalothorax is a little more enlarged than in the species
from the " Ogoue," as is readily shown by the measurements.
The upper surface is somewhat less depressed, the branchial regions
are somewhat swollen, especially the anterior ones, and the gastric
and cardiac areas are, in aged individuals, also slightly convex and
not so much depressed as in P. africanum. The oblique furrows
or depressions limiting off" the protogastric areas from the anterior
branchial lobes are scarcely distinguishable in P. africamim ; in
the other species, however, they are quite distinct, and, in adult
individuals, rather deep. The urogastric lobes and the cardiac
area are a little hroader in proportion to the breadth of the
carapace than in P. africanum. In the latter the gastric region
is faintly granular or rugose just behind the postfrontal ridge,
but in the new species it appears everywhere smooth, though
finely and sparsely punctate. The anterior branchial area is
slightly rugose in P. africanum., but not marked with oblique
wrinkles ; these are quite distinct in the species from the River
Prah and Liberia ; on the posterior branchial lobe these rugosities
occur in both species, they are, however, thinner and finer in
P. latidactylum. The posterior part of the upper surface

PZ.S. 19 03, vol. I. PL IX.

\i&

y IMS

•#

^^""

J. G.deMan del.

TAMON ( POTAMONAUTESJ) LATIDACTYLUM.

^-I 'OLJ

1903.] FEESUWATEE CEAB FEOM UPEEE GUINEA. 43

appears, on either side of the cardiac and intestinal regions,
somewhat granular and rugose in the Congo species, but^'^^e
smooth, though finely punctate, in P. laiidactyliim.

In the large aged female from the River Prah the oblique ruga3
near the lateral boundaries of the carapace show a tendency "to
disappear.

In both species the postfrontal ridge is rather prominent,
smooth, and nowhere granulated. In the type specimen of
P. africmmm it extends in a nearly straight hne toAvards the
second epibranchial tooth, uniting with that of the right side,
but ending about 1 mm. short of that on the left. The postfrontal
ridge of P. latidactylum usually curves slightly forwards mesially
and at each end ; it never unites with the epibranchial teeth,
ending about 1^ or 2 mm. short of the middle of the first epi-
branchial tooth. It is situated a little more distant from the
orbits than in P. africanum., so that the furrow behind the orbits
is somewhat narrower in the latter species. The front has the
same form and breadth in both, and shows in the middle a broad,
though shalloio hay ; both the front and the furrow behind the
orbits are a little granular in P, africantmi, but quite smooth,
though finely and sparsely punctate, in P. latidactylum..

The extraorbital and the two epibranchial teeth have a qidte
different form (PI. IX. figs. 1-3 & 7, 8). In P. africammn
(PI. IX. figs. 7, 8) they are more jwominent and separated from
one another hy much deeper incisions. The first epibranchial
tooth is, in this species, a little smaller than the extraorbital, but
has about the same form ; its straight or slightly arcuate outer
margin makes a right angle with the anterior margin of this
tooth. In P. latidactylum, however, the extraorbital tooth and
the anterior epibranchial one are much less prominent and mttch
loiver, the anterior margin of the first epibranchial tooth being
considerably shorter than its outer margin, measuring only one-fifth
of it. In the aged female from the Eiver Prah the extraorbital
tooth appears somewhat longer than the epibranchial one, and in
the two males (PI. IX. fig. 1) they have about the same length;
in the much younger female from Liberia (PL IX. fig. 3) the
epibranchial tooth, however, is almost twice as long as the extra-
oi-bital, but in the quite young male from the same locality
(PI. IX. fig. 2) the extraorbital tooth appears again once and a
half as long as the epibranchial. As regards the relative length
of these teeth, P. latidactylum presents therefore considerable
individual difierences. In both species the second epibranchial
tooth passes backward into an arcuate crest, bounding the cephalo-
thorax later;il]y ; this crest, which in young individuals of P. lati-
dactylum (PI. IX. fig. 3) appears distinctly denticulate, the den-
ticulations or crenulations disappearing in more aged individuals,
reaches in P. africanum to th^ 2^osterior boundary of the tiro-
gastric lobes and does not curve imoard on to the surface of the
branchial region : in P. latidactylum, however, it is much shorter,

44 BR. J. G. DB MAN ON A NEW [Jnil, 20,

scarcely reaching to the level of the mesial crescentic portion of
the cervical suture, and it distinctly curves inwards on to the
upper surface of the carapace. The orbits have the same form.
in both species, but in P. africanuin the incision that separates
the extraorbital tooth from their lower margin is deep, much
deeper than in the new species, so that in the former the outer
angle of the orbits strongly projects beyond this margin.

In P. africanum the transverse furrow, limiting off the sub-
hepatic region from the branchial floor, is bordered by a row of
small granules, the subhepatic area is covered with prominent
rugosities and granules ; in our new species the latter appears
almost smooth, and the row of granules is also less distinct. The
rugosities on the inflected portion of the cephalothorax and on
the outer part of the pterygostomian regions are also much less
distinct in the species from Upper Gviinea than in P. africanitm.

The outer foot-jaws have the same form and charactei's in both
species, and the furrow on the ischium-joint runs in both a little
closer to the internal than to the outei- margin of this joint.

The type of P. afo^icanimi is a female, the abdomen of which
has not yet obtained its full development and size. The sterniim
of P. latidactylum is smooth, punctate, and in both species a
transverse furrow unites the postero -external angles of the buccal
frame. The male abdomen (PI. IX. fig. 4) somewhat resembles
that of P. consohrinum de M. (Notes Leyden Musevim, 1899,
pi. 10. fig. 10,?"). The lateral margins are slightly concave. The
terminal joint has about the same length as the penultimate ; it
is triangvilar, rather pointed at the tip, not rounded like that of
P. consohrinum ; the lateral margins are veiy slightly arcuate
posteriorly, nearly straight towards the tip. The posterior margin
of the penultimate joint is a little broader than the anterior ; the
lateral margins, slightly convex anteriorly, are a little concave
posteriorly. The length of the antepenultimate joint measures
about two-thirds that of the penultimate, its posterior margin is
a little concave, so that the lateral margins are somewhat shorter
than the length of this joint in the middle line. The lateral
margins of this and of the following fourth joint are sti'aight, but
those of the fifth are rounded ; in P. consoh-inum, however, this
latter is also straight. As regards the coarse punctation, P. lati-
dactylum agrees with P. consohrinum.

In the two males from the River Prah the right chelipede is a
little larger than the left ; in the very young one from Liberia
the left is a little larger. The upper margin of the arm of the
larger male from the River Prah bears some transverse i-ugosities,
and one observes a small tubercle on the concave upper surface at
the proximal end, somewhat nearer to the upper than to the
anterior margin ; the latter is also somewhat tubercular on its
proximal half. The anterior surface of the arm bears, near the
articulation of the wrist, a conical tubercle not far from the
anterior margin, and a lower, broader tubercle between it anr'

1903.] FRESUWATER ORAB FROM rPPER GUIXEA. 45

the under margin; the latter is also sh'ghtly tubercular alonc'
its distal lialf. The three faces of the arm are quite smootli.
The carpus is also nearly smooth above, though spai-sely and
rather finely punctate ; the inner margin is armed witli a i-ither
small pointe<l spine in the middle and beneath it with a small
acute tubercle. The larger chela (PI. IX. fig. 5), measured
horizontally, is a little longer than the length of the ce})halotliorax
and the fingers are a little longer than the palm ; the latter is
just as long as high near the ai-ticulation of the fingei-s. The
convex outer suiface of the palm is quite smooth, though finely
punctate, similar to the upper margin, but the lower edge of the
palm is ol)soletely tubercular. The fingei-s are someichat com-pressed
especialh/ the rmmobile one. I'he immobile finger is ratlier hiqh
at the base; \ti^ flattened outer surface appears minutely and
densely granular under an ordinaiy lens, and one observes on it
three longitudinal rows of impressed points running to the tip of
the finger, the middle one of which is situated a little closex- to
the lower margin of the finger than to the ujipei-, and looks
like a very shallow furrow. The 15 or 16 teeth are small, two
or three are somewhat hu'ger than the others, and one in the
n\id<lle of the finger is the lai-gest of all. The fingers leave a
narrow interspace between them when closed, the pointed tips
being only in contact and crossing one another. The dactvlus is
somewhat arcuate and tapers rather slowly to the tip ; both on
the outer and inner sui'faces and on its upper edge this finger
appears minutely gramdar under a lens, and one observes on it
several rows of small shallow puncta. The teeth agree in size and
in number with those of the index, three or four being somewhat
larger than the others. The little convex inner surface of the
palm is quite smooth.

The left chelipede is somewhat smaller, the hand beino- 41 mm.
long ; it fully agrees with the right one, but the fingers are in
contact and the I'ows of puncta ai'e a little moi-e distinct.

In the young male from Liberia the fingers are in contact in
both chela?, the upper margin of the palm appears distinctly
granulate under a lens, and the longitudinal furrows on the
fingers are already visible to the naked eye.

In the young female from Liberia the right chelipede is a little
larger than the left, the fingers are in contact, and the furrows
on the index, as also the rows of puncta on the mobile finger,
are distinctly visil)le to the naked eye. The upper margin of the
palm of the left hand appears finely gramdar.

In the aged female from the River Prah the left chelii^ede is
much larger than the right ; the hand, indeed, is 45 mm. long,
just as long as the cephalothorax, and 20,| mm. high, but the
right chela (PI. IX. fig. 6) is only 31 mm. long and 13 mm. high.
In both the fingers are in contact throughout their whole len<>1:h
and the pointed tips cross one another ; the strongly compressed
and much flattened immobile finger is not furrowed, though the

46 DR. J. G. DE MAN ON A NEW [Jan. 20,

rows of puncta are still visible. The fingers are but little longer
than the palm, and the latter is on its outer surface and on its
upper margin smooth, sparsely punctate. The fingers of the
smaller chela, however, are once and a half a,ii long as the palm
(PI. IX. fig. 6).

The female type-specimen of P. africanum compared with the
female from Liberia, which is about the same size, shows the
following differences : — The anterior legs (PI. IX. fig. 9) are equal,
as^ regards their shape and size. The transverse rugosities on
the upper margin of the arm are more distinctly granular, and
the anterior and lower margins are also more distinctly tuber-
culated ; the conical spine on the anterior surface, near the articu-
lation of the wrist, is more pointed, and several small tubercles
are seen between it and the anterior margin of the arm, that are
wanting in P. latidactylum. The upper surface of the carpus is
slightly granular at the base of the spine on the inner margin,
and one observes, moreover, everywhere fine impressed lines and
points ; the spine is somewhat larger than in P. latidactylum,
and instead of a small tubercle, there is beneath it a smaller,
pointed spine. The hands (PI. IX. fig. 9) have a quite different
form. The fingers, almost in contact, are distinctly somev:hat longer
than the palm, and the latter is distinctly less high than long. The
upper margin of the palm is covered with depressed granules,
and the puncta on the convex outer surface are partly arranged
in longitudinal rows. The fingers are not compressed, slender,
7ior curved, except at the pointed tips which cross one another.
The immobile finger is much loioer at its base than in the
other species, and its convex outer surface is distinctly furroived ;
of the three furrows visible to the naked eye, the middle one is
deeper than the others. This finger is armed with 25 or 26 small
teeth, some of them on the pi'oximal half are somewhat larger
than the others. Finally, the dactylus tapers regularly to the tip
and is also longitudinally furrowed by rows of impressed points ;
the teeth agree with those of the index. The fingers are almost
smooth.

The ambulatory legs, the measurements of which are giA^en
on p. 47, apparently agree with those of P. latidactylum, but the
meropodites are -inore granular along and near their upper edges.

Potamon auhryi H. M.-E., from the Gaboon, P. pelii Herklots,
from the Gold Coast, and P.floweri de M., from the Soudan, are
different species (mcZe de Man, Proc. Zool. Soc. Lond. 1901, vol. i.
p. 94), being at first sight distinguished by the upper surface of
their cephalothorax being very convex from behind forwards,
smooth and shining, hy the different form of the fingers, of the
abdomen, &c. The young male of Potamon pelii described I. c.
p. 99, I have re-examined for the purposes of this memoir.

1903.] FRESHWATER CRAn FROM UPPER GUIXEA.

Jleasurements in millimetres.

47

1.

2

3.

?.

<?.

<?.

67

58

51

45.V

39

3o

40.V

35. V

31

57i

50

4^^-

18

16

14

6

5

4i

5i

4^

4$

10

8*

6f

11

8.V

7

17

12

lo.v

11

9i-

8i

6

o'r

o

7

5t

6.V

6

8i

7^

10.',

9

45

46

39

21

221-

19i-

20.V

22?,

19^

80

76

71

27

25

23^

Si-

8.!r

n

lo

13

m

6.1

6

oi-

20.^

17

ls.',

72

61

57

21{-

191

18

8

7.V

7

lU

lOi

n

7

6i

6

17

12i

11

<?.

Breadth of the cephalothorax

Length ot'tlie cephalothorax without the uhdomen...

Distiince between the external ovliitiil angles

„ „ posterior eiiihraiichial teeth ...

Breadth of tho anterior margin of the front

Distance, in tho niwlian line, between the anterior

margin of the front and the postfrontal ridge

Distance between the outer angle of the orbits and

t he post frontal ridge

Distance between the outer angle of the orbits and

t he posterior epibranchial tooth

Breadth of the iirogastric arciu

„ „ cardiac areai

,, ,, orbits

Height of the orbits, near the lower internal angles .
Length of the terminal segment of the abdomen

,. ,, penultimate segment

Breadth of the anterior margin of this segment

,, ,, posterior margin of it

Length of the larger chela

„ „ palm

Height of the palm, near the articulation of the

dnctylus

Length of the legs of the penultimate pair

„ „ meropodites of the penultimate pair...

Breadth „

Length of the propoditcs „ „ „ ...

Breadth „ „ „ „ „ ...

Length of the dactylopodites ,, „ „ ...

„ „ legs of the last pair

„ „ meropodites of the last pair

Breadth „ „ ,, „

Length of the propodites „ „

Breadth „ „ „ „

Length of the dactylopodites „ „

48i 34J
33 24|

my. 23

42.1
14

4

n
65
10

16

15i

66

21

8
12

b\
151
51

IQk

7

91

6
101

3U

loi

3i

H

71
6i
3f

^\
4-1

22i 34
11 116

13

51
7

11

50

161

6i .
91 15.V
4 Q\

111 j 18

39 60

20

8
12

7^
lU

Nos. 1-3, River Prah ; Nos. 4, 5, Liberia; No. 6, type-specimen of Theljiihusa
africana A. M.-E., Ogouc, from the I'aris Natural History Museum.

EXPLANATION OP PLATE IX.

Figs. 1-6. Potamoii (Potamnnautes) latidacti/him, n. sp. — Fig. 1. Dorsal view of
the anterior part of the carapace of the largest male specimen from tho
River Prah, X U-. Fig. 2. Left antero-external angle of the carapace of
the young male from Liberia, X 3. Fig. 3. The same of the young female
from Liberia, X 3. Fig. 4. Abdomen of the largest male from the River
Prah, X 11. Fig. 5. Larger chela of the same male from the River Prah,
X 11. Fig. 6. Smaller chela and carpus of the adult female from the
River Prah, X 11.

Figs. 7-9. Potamon (Potamonautes) africamim A. M.-E., female type-specimen of
the Paris Jfuscum, from Ogoue.— Fig. 7. Dorsal view of the anterior part
of the cephalothorax, X 11. Fig. 8. Left antero-external an-le of the
carapace, X 3. Fig. 9. Right chclipcde, X l-V.

48 MR. R. I. POCOCK AND HON. N. C. ROTHSCHILD ON [Jan. 20,

6. On a new " Bird's-dung ^' Spicier from Ceylon. By R. I.
PococK, F.Z.S., and the Hon. N. C. Rothschild, B.A.,
RE.S., F.Z.S.

[Received November 18, 1902.]
(Plate X.^)

During a recent visit to Ceylon, the Hon. N. C. Rothschild, in
company with Mr. E. E. Green, discovered a specimen of a
Spider (Phrynarachne) imitating, for purposes of allvirement and
concealment, a patch of bird's-dung. A photograph of the spider
resting upon the leaf iii situ was obtained, and the leaf with its
patch of web was brought home with the spider (PL X. figs. 1, 2),

The film of white web upon the leaf, with its irregular outline
and marginal, often apically thickened projections, closely assimi-
lates the splash of the fluid components of the excrement ; while
the spider itself, with its mottling of black and white, resting
upon the centre of the film, with its legs tucked in, represents
the semi -solid central core. So close, indeed, was the imitation
that both observers were for a time deceived, until closer scrutiny
revealed the true nature of the deceptive patch.

It is commonly assumed that the purpose of this imitation is
purely alluring or pseudepisematic. ISTo doubt, however, it is an
equally impox'tant pi'ocryptic factor, serving to protect the spider
from various enemies, especially the wasps of the family Pompilidse.

The spider, which appears to represent an undescribed species,
may be named after one of its discoverers ^.

Phrynarachne rothschildi, sp. n.

2 . — Colour. Carapace with a broad central black patch, with
irregular lateral margin, extending from the eyes to the posterior
slope, and forming an angular expansion on each side imme-
diately behind the head ; and with its posterior border passing
into a median longitudinal stripe, continuous with a trans-
versely arcuate black stripe which extends from the middle of
the posterior slope of the carapace on to its lateral portion,
breaking up into fainter patches separated by radiating pale
bands ; clypeus, sides of head, and area between central patch and
lateral patches yellowish white. Mandibles yellow in basal half,
with a large black patch, not reaching the fang, in the distal half
in front, blackish brown benea,th. Sternum mostly yellowish white,
with a large subtriangular black patch, di-\dded by a narrow pale
line in its posterior half, and some black at the bases of the legs.
Labium blackish, maxillfe blackish internally, whitish externally.
Palp with its femur white below, black above in its basal half ;
patella white, tibia and tarsus black, variegated with yellow
patches. First and second leg's with the coxae black, variegated

1 For explanation of the Plate, see p. 51.

2 Mr. R. I. Pocock takes the sole responsibility for tlie naming and description of
the species.

p. Z. S. 19 03^ vol 1. PI X

F.Hckard- Cambridge del eftlith,

PHRYNAPAC HNE RO THS CHILDI

West.Nevnnan imp.

1001).] A Xi:w SPIDKU KIIOM CEYLOX, 49

with dirty yellow below, whitish witli a black spot above ; femora
jet-black variegated with paler patches alx)ve, in front, and
behind, with a uairow whitish line between the tubercles below,
the distal end yellowish white; patellar yellowish white, with
two palely fuscous patches above and some behind ; tibise and
protarsi mostly jet-black, with a, pale median inferior longitudinal
stripe between the spines; tarsi pale above, black below; third leg
with coxa and ti-oehanter blackish below, the femur amber-yellow
in its liasal two-thirds, yellowish white with a black [)atch, much
broader liehind than in front, in its distal third ; tibia black and
whitish, the l)lack predominating posteriorly; protarsvis and tarsus
mostly yellowisli; fourth leg, with exception of its coxa, mostly
yellowish whit(% lightly spotted above. Almost the whole of the
upperside of the abdomen jet-black from its anterior margin
backwards to the spinners, with the postero -lateral tubercles and
the sides yellow ; the under surface {i. e. the area between the
lung-sacs on the epigastric region and behind the genital fold
nearly back to the spinners) black, this black fielil showing a ti'ans-
vei'sely oblong pale patch behind the genital apeiture (Plate X.
figs. 3 & 4).

Carapace not strongly tubercular, flattish longitudinally along
the middle line, with a shallow transvei'se depression behind the
ocular area of the head ; area between the ocular tubei-cles
depressed ; eyes of posterior line (fig. 5) subecpially spaced, the
medians about 5 diameters apart, lying about their own diameter
in front of a tangent joining the anterior borders of the lateials,
which are distinctly largei- than the medians ; eyes of anterior line
(fig. 6) strongly recurved, the inferior edges of the latei'als as high
as the superior edges of the medians, the latter considerably smaller
than the laterals, about 2.| diameters from each other and about
2 diameters from the laterals ' ; distance between the medians
about equal to the height of either above the edge of the clypeus,
which is weakly tiibercular and furnished with ^n'orainent angles.

Mandihles prominent basally, scarcely tubercular ; margins of
fang-groove thickly fringed, the posterior aimed with one fang, the
anterior with teeth, remote from base of fang; the posterior fringe
continuous with a scanty band of hair passing up the inner side
of the posterior surface of the mandible.

Legs. Femur of 1st leg thickly covered in front with small
tubercles, with a few large tubercles above and two rows of about
five tubercles each below, the posterior surface smooth ; tibia
bowed, armed beneath with two rows of about 4-5 spines, the
anterior and dorsal sin-face also spined ; protarsus a little longer
than tibia, armed beneath on each side of the middle line with a
band of longer and shorter inegularly arranged spines, those on

' Owinp to ;i want of dotinitinn of tlie bordov of the conipal Ipns, fhi^ pvocise ';i;!e
of the eyes of the antprior line, mid consequently of the width of the intcrsjiiice
between them, is hind to iisrertnin, varying: apparently in difli'vint lights and with
Ifnses of ditterent power. In the ahove-priven description the eyes are described as
seen under a 4-inch objective and a platyscopic lens.

Proc. Zool. Soc. -1903, Vol. I. Xo. IV. 4

50 MESSRS. R. I. POCOCK AND N. C. ROTHSCHILD ON [Jan. 20,

the posterior side being much less numerons than on the anterior,
the anteiior and dorsal surfaces of the segment also spiny ;
tarsus covered with rows' of spinules ; 2nd leg like the 1st, but
with the femur almost quite smooth ; 3i'd and 4th legs with tarsi
bristly beneath, tibia of 3rd spined below, of 4th scarcely spined.

Abdomen as wide behind as it is long, and twice as wide as it is
in front, the anterior border ti'ansvei'se and armed with four
tubercles, the lateral border with four small tubercles, the
postero-lateral angle with two large superior and two small
inferior pale tubercles ; between the former are four pairs of
shining black tubercles, tAvo pairs of which are mvich larger than
the others ; in addition to these the dorsal surface is furnished
with three pairs of small scattered tubercles ; lateral surface
pitted ; inferior surface with two rows of muscular pits. Vtdva
undeveloped.

Measurements in mm. — Total length 9, carapace 4 ; posterior
width of abdomen 5'5, anterior width nearly 3 ; length of 1st leg
14, its femm^ 4, patella + tibia about 5, 3rd leg 6, tarstxs + pro-
tarsus about 5.

Log. Kandy.

In size and shape this species does not appear to differ appreci-
ably from F. ceylonica 0. P. Cambridge (P. Z. S. 1884, p. 201,
pi. XV. fig. 3, sub Ornithoscatoides), which was recorded from
Ceylon. The colouring of the two, however, is very different.
In F. ceylonica the anterior legs are heavily blackened only upon
the protarsi and distal half of the tibige, the mandibles have no
black patch, and the black-and-yellow pattern of the carapace
and abdomen, so conspicuous in P. rothschildi, is but little differ-
entiated ; also the cephalic tubercles between the two rows of
eyes are much higher in F. ceylonica (see also Simon, Hist. Nat.
Araign. i. p. 1043, fig. 1087, 1895).

Another Ceylonese species, 0. nigra 0. P. Cambr. (Joe. cit.
p. 202, pi. XV. fig. 4), was based upon the male sex, and cannot,
therefore, be compared with those based upon females. A third
species, F. fatalis O. P. Cambr. (P.Z. S. 1899, p. 525, pi. xxx.
fig. 7), from Ceylon, differs so markedly in form, coloration,
tuberculation, &c., as to need no comparison with this new species.
Also F. peeliana Stol. (J. A. S. Bengal, xxxviii. 1869, p. 229,
pi. XX. fig. 4), from Sibsagar, Assam ; F. Uiberosa Blckw. (Ann.
Mag. N. Hist. (3) xiv. p. 38, 1864, & P. Z. S. 1884, pi. xv. fig. 2),
from the East Indies ; and the Burmese foi'm F.papiolaia Thorell,
are all different from this new form.

To Dr. H. 0. Forbes, F.Z.S.', belongs the credit of the discovery
that the coloration of the species of Fhrynarachne belongs to the
pseudepisematic category. The pattern of yellow and black which
decorates the dorsal and ventral sides simulates that of the semi-
solid central portion of a pa,tch of biixl's-dung splashed upon a leaf,
the paler more fluid portion being represented by a thin irregular-
shaped carpet of white silk, in the centre of which the spider
takes its stand. The spider discovered by Messrs. Ptothschild
1 P. Z. S. 1883, p, 586, pi. li.

1903.] A XEW sriDE!i rnoM ckylox. 51

and Green was resting upon the leaf in tlie uunnal position, tliat
is to say back uppermost (figs. 1 &> 2).

Forbes, however, expressly states that F. decipiens, the species
he discovered, lies back downwards on the wel), holding itself
in place by means of the sjiines with whicli the anterior upper
surfaces of the legs are furnisheil, and he adds that the under-
side of " its rather irregidarl y-shaped and prominent aljdomen
is almost all white, of a pure chalky white" (p. 587), and that
" its piu-e white abdomen represents the central mass of the bird's
excreta, the black legs the dark poi'tion of the slime." Refer-
ence, howevei', to his figure, which was apparently drawn partly
from memory, the leaf and web having " gone astray " during
the transport home, shows that the iniderside of the abdomen, so
far from being almost all white, is furnished with a large black
sub-oblong patch, which extends from its anterior boi'der con-
siderably past the middle. It is noticeable, too, that from the
ventral aspect the abdomen is not iiTegularly shaped, but evenly
oval in outline and devoid of tubercular excrescences. The dorsal
surface, on the contraiy, as repi'esented in the figui-e given by
INIr. Cambridge (P. Z. S. 1884, pi. xv. fig. 1), might very well
be described as almost all white and rather irregularly shaped.
The white colour largely predominates, and the tubercles project
prominently fi-om the expanded posterior portion. ISIention is
made of these discrepancies between the figure and description,
to justify the suspicion that Di*. Forbes may have mistaken the
dorsal for the ventral surface of the spiders he saw. Moreover,
one cannot but wonder how the spider maintains a secure hold
back downwards, especially when the powerful prehensorial legs of
the first and second pairs are released, as released they m\ist
surely be, to seize an alighting butterfly. One would think that
the flapping of the insect's wing would pull the spider, now
insecui-ely anchored, from its hold and bring both to the ground
together. On the other hand, if the spiders of this kind always
rest in the normal position seen and photographed in the case of
F. roihschildi, the simulation of the bii'd's-di;ng is equally perfect,
and the spider, while seizing a butterfly with its fore legs, can
maintain itself securely in place by gi'asping the web with the
claws of the remaining pairs and by gluing its spinning-mamilke
to the subjacent silken threads.

These, then, ai-e the points for further investigation which we
would ask i-esidents in the tropics to take note of : — Do the species
of Fhrynarachne assume indifl:erently the dorsal or the ventral
attitude ? Is it a peculiarity of one species to lie back uppermost,
and of another belly uppermost ? If they lie, as Forbes asserts,
back downwards, how is a secure hold maintained when an insect
has to be seized ?

EXPLANATION OF PLATE X.

Fhrynaraclviie rothscJiildi, sp. n., p. 48.

Figs. 1, 2. Tlie Siiiilcr vesting on leaf. Fijr. 3. Up])er suiiacc.

Fi^-. i. Under surface. Fics. .i, 6. Eves.

4*

52 DK. H. J. HAXSEN ON CRUSTACEANS [Jan. 20,

7. On the Crustaceans of the Genera Petalidhim and Sergestes
from the ' Challenger/ with an Account o£ Luminous
Organs in Sergestes cliallengeri, n. sp. By Dr. H. J.
Hansen (Copenhagen) .

[Received November 29, 1902.]
(Plates XI. & XII.')

Dui'ing a stay in London in July and August, 1902, I examined
various groups of Crustacea in the British Museum (Natural
History). I beg the Director, Professor E. Ray Lankester,
and Mr. F. Jeffrey Bell to accept my sincere thanks for the free
use of the collection and for their kind help.

In the paper " On the Development and the Species of the
Crustaceans of the Genus Sergestes " (Proc. Zool. Soc. Lond.
1896, pp. 936-70) I have given a revision of this extensive
genus. I had studied a very rich material of pelagic forms
belonging to the Copenhagen Museum, among which are all the
types of Kroyer ; besides I had examined types of 5 species
established by Chun, Metzger, and Ortmann.

Among other things, I proved that " of the 59 {or 60) hitherto
described species only about 20, or one-third of the total number,
have been established on adult animals, such as have almost or
entirely arrived at sexual maturity ; and that almost all the other
species are true larvce, and even of these a considerable mmiber are
larval stages of species already established on adult specimens, . . ."
Of earliei- authoi-s, C. Spence Bate has produced a very large con-
tribution on the genus Sergestes, extending to eighty-eight quarto
pages and seventeen plates, in his " Report on the ' Challenger '
Macrura." He established the genus Petalidium on a new species,
described 24 new species of Sergestes &c. In 1896 I wrote
(p. 939) : " This large contribution is of course of great
importance, but unfoi'tunately neither the descriptions nor the
figures are so good as could be wished, and in numerous instances
... a re-examination of the type specimens is absolutely necessary
— the greater part of the new species are but larvas." I have now
studied all types which are preserved in the British Museum, and
the present paper contains the results of my examination.

Bate describes 31 species of Sergestes as examined by himself :
of these, 24 are established as new to science, 6 are considei-ed
to be Kroyerian species, and one is referred to S. atlanticus
H. M.-Edw. The types of 9 of the species established by Bate do
not exist in the British Museum ; some specimens mentioned in
his work and belonging to other species are also absent ; but
several specimens belonging to varioiis species and omitted in the
Report were found in the collection. I am therefore only able
to give more or less incomplete notes, based on the study of the

1 For explanation of the Plates, see p. 78.

PZ S 1903, vol. I. PI, XI

H. J. Han.
FIG

j.y Edwir. Wilson , Camhnt/ae

l.PETALIDIUM FOLIAGEUM. 2.PETALIDIUM JUN. 3 SERGESTES
PROFUNDUS. 4. S.PREHENSILIS 5. S.KROYERI 6. SSIMILIS.

P.ZS. 1903, vol. I PI. XII

del. Edwin Wilson. Cambridge

FIG.-l.SERGESTES ARCTICUS. 2.S.GHALLENGEHI .

1903.] IN- THE ' CnALLEXGER ' COLLECTIOX. 53

specimens, on 22 of the species in question, 15 of which were
established by Bate himself. Furthermore, he refers Petalidium
Bate, Sciacaris Bate, Acetes H. M.-Edw., and Lucifer Vaugh.
Tliomps. to his family Sergestidte ; of Lucifer he describes two
species, but the genus must be more thoroughly studied than has
hitherto been done before the examination of the ' Challenger '
specimens ; Sciacaris Bate has one species, which is only a larva of
a Sergestes, and the tyjje seemed to be wanting in the Museum ;
of Acetes, Bate had no specimens; and Fetalidium is mentioned
below.

Of Bate's 31 species of Sergestes only 6 arc really mature forms,
25 being larvie. Special attention has been jjaid to the adult speci-
mens j)i'esei'ved in the Museum and enumei-ated by Bate ; on two
of these specimens I have established two new species, and
besides I add notes and some drawings to the representations of
Bate. Unfortunately nearly all the specimens of rare and
especially interesting species were veiy much mutilated.

Our pi'esent knowledge of the adult species of Sergestes of the
Atlantic fauna is far from complete, liut yet we ai'e acquainted
with so many species that it was possible for me in my earlier
paper either to refer the Mastigopus-iovYL\^s, examined to the
mature species, or to describe the older larval stages and some-
times the black-eyed but still immature forms, so that they can be
recognized with certainty and referred to the mature forms when
these are discovered in the future. All the Atlantic lai-val forms
from the ' Challenger ' seen by me have now been referred in a
similar way. But many lai-vse established by Bate as valid species
of adult or sub-adult animals have been secured in the Pacific.
Our knowledge of the mature stage of the species living in that
vast ocean is still rudimentary ; and I have therefore not been
able to refer the lai-vfe of three of Bate's species to any species
established on adult specimens. Bate's types of his species
established on larval forms are often either defective or very
young, wherefore I thought it of little use to describe and figure
them again ; but I have generally added some notes on their
affinities, and sometimes also a few corrections to his descrip-
tions. "When the Pacific has been modei'ately well explored
by further expeditions, many adult forms and their larval stages
will be discovered ; and a future student of the group will then be
able to refer at least some of the larva?, which I cannot interpret,
to their adult foi-ms. To the young larvse described b}' Bate as
species of Mastigopus I pay no attention at all : the types seem
to be lost.

I think it convenient first to deal with all the ' Challenger '
species in the Siime consecutive order in which they are described
in Bate's Report ; then to put together some residts of the investi-
gation ; finally, to mention more fully tlie luminous organs in
Sergestes chcdlengeri, n. sp.

In order to abridge the descriptions, in the following pages —
as in my earlier paper^ — I make use of some abbreviations : —

54 DR. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

antenn. ped, = peduncle of the antennulse ; mxp.^-mxp.^ = the
second and third pair of maxillipeds ; trl.^-trl.° = the first to the
fifth pair of trunk-legs ; br. = the first, br.^ = the second branchia
above the same legs ; ext. br. of urp. = external biunch of the
tiropods.

I. Notes on the Species of Petalidium and Sergestes.

a. Petalidium Bate.

To this genus Bate has referred only one species. It will be
convenient to deal with the characters of the genus together with
those of the species.

Pet. foliaceum Bate, pp. 348-50, pi. Ix. (Plate XI. figs, la-1 g.)

Bate mentions five specimens from two stations : Stat. 146,
lat. 46° 46' S., long. 45° 31' E., 1375 fathoms; and Stat. 159, lat.
47° 25' S., long. 130° 22' E., 2150 fathoms. All these specimens
are at present in the Museum, but even Bate's text and his figure
of the entire animal show that most of the appendages presenting
specific characters are wanting or have been mutilated. Bate
writes (p. 349) : " The great distinction between this genus and
Sergestes exists in the form, character, and arrangement of the
branchial plumes, which consist of a series of plates and cylin-
drical filaments, situated side by side in a series of rows at right
angles to the stalk. There is but one plume to each of the five
anterior somites of the j)ereion, the posterior two somites having
none ; between some of the somites is a large foliaceous plate."
But this description is difiicult to understand : his figure 3,
representing the branchiae, is defective, and his tabular view
(p. 349) is wrong. He was of opinion that the foliaceous plates —
of which he had seen only three — were pleurobi'anchise, Avhile the
"branchial plumes " were arthrobranchiae ; but this is incorrect:
they are decidedly pleiii-obranchiae as in Sergestes. Petal. foUaceimi
Bate dififers from all species of Sergestes in one quite unimportant
feature, that no trace of branchiae is found above trl.'*, and in one
important character, viz., the sti'ucture of the pleurobianchial
plumes. This structure is very interesting (fig. 1 e). The most
developed branchiae are, as usual, those above ti'l.^ and .trl.'^ ; each
of these consists of an anterior and a posterior half, and each half
of five (to six) transverse rows of branchial plates, generally five
or six in each row, and these plates (some of the lowest excepted)
are directed upwards. The pleurobranchiae above mxp.^ and
especially trl.^ are less developed, with a lower number of trans-
verse rows, and partially with a loAver number of plates in the
rows. The pleurobi'anchial plumes in Fetalidium look very
difiei'ent from those in Sergestes ; the real differences are : a
much loAver number of rows, a much lower number of plates in
the rows, and that the plates are much larger, curved upwards, and
look much more independent. Behind the upper part of each of
these four pleurobranchise originates a pleui'obranchial lameUa (Z.),

1903.] IX THE ' CHALLENGER COLLECTIOX. 55'

which is a I'educed branchia ; these lamellae are very long above
trl.^ and ti*l.^, while the two anterior lamellse above mxp.^ and
trl.^ are much less prominent. On mxp." is found an epipod (e/?.),
with a branchia consisting of a few plates, and above it a rudi-
mentaiy pleurobianchia consisting of one very small lamella.

Fig. 1 a and fig. 1 c show the rostrum, consisting of a carina
with a short or veiy shoi-t terminal spiniform pi'ocess. There are
no supraocular or hepatic spines ; the gasti'o- hepatic groove is
well developed. A comparison of fig. 1 c with fig. 1 d shows that
the eye-stalks are considerably depi'essed ; they are from two and
a half to neai-ly three times longer than the eje& at the inner
terminal angle, with a small rounded knot turning inwards and
a little upwards (fig. 1 h ; fig. 1 d), and besides (always ?) with an
exceedingly small tubercle somewhat in front of the inner angle
and more downwards on the inner side (fig. 1 d). In the antenn.
ped. the basal joint is very broad (fig. 1 &), decidedly shorter than
the outer margin of the two other joints together ; the third joint
is scarcely three times longer than broad, considerably longer
than the outer margin and a little shorter than the inner mai-gin
of the second joint, which is stout, with its inner margin only a
little moi'e than twice as long as its breadth. (The sjoecimen
from which figs. 1 c and 1 e were drawn measures 51 mm. in
length, and was captured on Stat. 146 ; figs. 1 a-\ h were drawn
from the specimen secured on Stat. 159).

But besides these five specimens of FetaUdhtm, I found still
two smaller specimens of the same genus among the ' Challenger '
animals. One of these had been determined h.s 6'ergestes japonicus
Bate, but is not referred to that species in his text. It was
captured at Stat. 158, 7/iii., 1874, 1800 fathoms, thus near one
of the above-named stations. It measures only 21 '5 mm. in
length. Figs. 1_/ and 1 g show that its I'ostrum has the terminal
process somewhat longer than in the large specimens ; the eyes
are a little longer as compared with the length of their stalks, and
these are proportionately somewhat broader, without a distinct
knot at the end of the inner margin. There is no supra-ocular
spine, but the hepatic spine is moderately developed ; the gastro-
hepatic gi-oove is very distinct. No branchiae above trl.^ This
small specimen belongs certainly to P. foUaceum Bate.

Bate established (pp. 428-31) his Sergestes profundus on two
specimens, both badly mutilated. He describes each specimen
sepai'ately, beginning with one captured at Stat. 137, lat. 35° 59' S.,
long. 1° 34' E., dej)th 2550 fatlaoms. But, unfortunately, this
specimen does not belong to Sergestes but to Petalidium ; the
type is besides so mutilated that I should have preferx'ed to
omit it, if it had not been described by Bate. It measures
17"5 mm. in length. The rostrum is shown in fig. 2a; it is
described by Bate : "It consists of a short fine point projecting
horizontally for about one-fourth the length of the ophthalmopod,
and is dorsally furnished on the crest with a small tooth." The
eye-stalks are as in the sicpJl speci/nen from Stat. 158 just .

56 DR. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

described ; the posterior brancliiiB to a large degree are destroyed,
and as to the other features, I refer to Bate's description (p. 429).
I cannot say with certainty whether the animal is a very young
specimen of P.foliaceum Bate, or belongs to an unknown form ; the
rostrum deviates considerably from the type of P. folkic&um, but
its shape presents a stage betv/een those in the adult and in a larva,
nearly agreeing with that in a larval form mentioned below.

In my earlier paper on Sergesies I described S. obesus Kr.
{ = S. sanguineus Chun). I stated that it was a larva, a Masti-
gopus, and added (p. 968) : " It is easily seen that this species
cannot remain in the genus Sergestes, but whether it should be
referred to Petalidium Ba,te, or a new genus should be established
for its reception, is impossible to decide with certainty. The
branchial plates recall the plates found in Petalidium, and there-
fore I provisionally transfer it to that genus. ..." I had not
seen any adult Petalidium, but suggested (p. 967) that the
branchial plumes intei'preted by Bate as arthrobranchise are in
reality pleurobranchise. I can now decide that S. obesus Kr. is
the Mastigojms of Petalidium : the branchia3 pi-ove it ; besides,
trl.^ in the Mastigopus is exceedingly elongate, with the basal
part very thick, and in the adult form the basal joint of the same
pair (fig. 1 e) is exceedingly thick, very much thicker than those
of trl.^ and trl.^. Finally, the shape of the rostrum of S. obesus
Kr, agi-ees nearly with that of the young Petalidiicm from
Stat. 137 just mentioned. I described the branchicB of &'. obesus
Kr., stating that in the largest specimen a rudimentary branchia
was found above trl.* ; in the largest but one of the specimens I
have now looked for this branchia, and found it. Finally, I stated
that P. obesttm (Kr.) is "decidedly distinct irom P. foliaceiim Bate,"
and I maintain this opinion, founding it on the fact that P. foli-
aceum does not possess any bi'anchia above trl.'' ; and it is very
improbable that such a branchia begins to develop itself in the
Mastigopus and disappears again in the adult.

b. Sergestes H. M.-Edw.

Serg. intermedius Bate, p. 383 (no figure).

Bate mentions one specimen, 5 mm. in length, from the
" China Sea, off Luzon." The specimen has been preserved ;
according to the rostrum, the eye-stalks, and the ext. br. of urp.,
it belongs to *' Serg. brachyorrhos Kr.," which is the youngest
Mastigopus of S. edivai-dsii Kr. (P. Z. S. 1896, pp. 963-64).

Serg. prehensilis Bate, p. 385, pi. Ixxi. (Plate XI. figs. 4 a,
ib.)

Bate has examined one sj)ecimen from Stat. 236, lat. 34° 58' N.,
long. 139° 29' E., off Japan, 775 fathoms. The type, an adult
male, has been preserved ; it differs from all other species known
to me. It will be useful to redescribe its most essential characters,
and two new figures are given (PI. XI. figs. 4 «, 4 6). The

1903.] IX THE 'challenger' collectiox. 57

ro.sti'um is rather long, directed forwards and considerably
upwards ; its terininul portion is jiroduced into a spiuifonu
piocess, and the upper margin of the rostrum has at the base of
that process a sliarp angle as a rudiment of a spine ; the lower
mai'gin of the rostrum between its base and the apical spine is
sti'ongly convex, the upper margin nearly straight. Supra-ocular
antl hepatic spines are wanting, the gasti'O-hepatie groove slightly
developed. The eyes are lai'ge, considerably depressed ; seen from
the side (fig. 4 a), they are somewhat longer than the whole stalk ;
seen from above, their basal margin is very oblique (fig. 4 6), so
that the interior margin of the distal joint of the stalk is as long
as the outer margin of the eyes. The antenn. ped. with the outer
margin of the first joint is a little longer than that of the two other
joints together, and only a little shorter than their inner margin ;
the second joint Avith the inner margin is three times longer than
the breadth, and somewhat longer than the third joint, which is
about two and a half times longer than broad. The antennal
s({uama reaches nearly to the end of the antenn. ped., with its
distal portion broad and the outer spine well developed. Mxp.^ is
a little shorter than trl.^ The l'ranchi;e present a ti'ansition-form
between those of b'. arciicus Kr. (PI. XII. fig. 1 c) and tS. robusivs
Smith ; the pleurobianchial lamella above mxp.^ is very small.
Of the bianchije above trl.^, br.^ is scarcely two-thirds as long as
br. ; of those belonging to trl.^, br. is slightly longer than br\
and br.^ a little shorter than br} above trl.^ The ext. br. of vu-p.
is almost five times longer than broad, and the haiiy portion of its
outer margin is a little more than one- fourth of the total length. —
Length 36"5 mm.

This species occupies an intermediate position between S. arcticus
Kr. and S. robitsius Smith. Bate's fig. 4, showing an antennide
and an eye, is misleading, the antennular peduncle being too
slender, with the basal joint too short, the third joint too long.

Serg. japoxicus Bate, p. 387, pi. Ixx. figs. 1, 2.

Bate enumerates three specimens from two localities : Stat. 232,
lat. 35° 11' N., long. 139° 28' E., 345 fathoms; and Stat. 207,
lat. 12° 21' N., long. 122° 15' E., 700 fathoms. All have been
preserved, and belong to one species. In 1896 I wrote that
S . j((j)onicus Bate must le identical with ^S'. mollis Smith (taken
in the Atlantic, ofi' the United States), and gave reasons for
my view. On comparing Bate's specimens with Smith's elaboi-ate
description (Rep. U.S. Comm. Fish & Fisher, for 1882, p. 419,
1884) and his figures (Rep. U.S. Comm. Fish etc. for 1885, pi. xx.
figs. 3-5), I ai-rived at the same lesult. It should be especially
mentioned that an examination of the bi'anchias showed the most
complete agreement with Smith's description and drawing. For
full information on S. japonicus Bate, I I'cfer, therefore, zoologists
to the papers of Smith.

As already mentioned, 1 found in the bottle with S. japonicus
from Stat. 232 a tube containing a smaller animal determined

58 DR. H. J. HANSEN ON CRUSTAOEANS [Jan. 20,

to be S. jajyonicus and labelled Stat. 158. It is not mentioned
in Bate's work. On a closer examination it turned out to be a
specimen of Fetalidium, and it lias been described above.

Serg. kroyeri Bate, p. 388, pi. Ixx. figs. 3, 4. (Plate XI.
figs. 5 a, 5 b.)

Bate established this species on one large specimen, from
Stat. 170, lat. 29° 55' S., long. 178° 14' W., 520 fathoms. The
specimen is very mutilated ; a new description with two figures
(PI. XI. figs. 5 a, 5 b) is here given.

The rostrum (fig. 5 a) is rather low, rounded above, with the
upper front angle blunt and slightly projecting and the anterior
niai'gin concave ; but it could not be settled whether the upper
margin of the rostrum had been damaged or presented its natuial
shape. Supra-ocular and hepatic spines are wanting, the gastro-
hej)atic groove is strongly developed. The eyes (fig. 5 b) are
large, somewhat depressed ; seen from above nearly as long as
broad, slightly longer than the outer, and decidedly, but not
much, longer than the inner margin of the stalk. On the upper
side of the stalk, close behind the eye and near the inner margin,
is seen a rather small, subcylindidcal, distally rounded process,
directed obliquely forwards, inwards, and upwards ; it seems to
terminate in an organ. In the antenn. ped. the basal joint is
broad, with the outer margin a little shorter than that of the
two following joints together, but only two-thirds as long as the
inner margin of the same joints ; the second joint is stout, with
the inner margin scai'cely three times longer than the breadth ;
the third joint is stout, but yet considerably more slender than
the second, scarcely three times longer than broad, somewhat
longer than the outer and somewhat shorter than the inner
margin of the second joint. The squama of the antennas seems
to be nearly as broad at the distal end as in S. japonicus Bate,
but it could not be seen whether the outer spine is developed.
Mxp.^ and all trunk-legs are wanting. The pleurobranchial
lamella above mxp.^ is small. The branchia3 above trl.^ have been
broken ofi" on both sides ; of the branchiji3 above trl.*, br. is nearly
three-fourths as long as br. above trl.^, while br.^ is slightly more
than half as long as the same. The ext. br. of urp, with the
apical part is wanting ; the branch seems to have been at most four
times longer than broad, perhaps without marginal spine, and with
the haired part of the margin unusually short. — Length 60 mm.

The species is allied to t^. prehensilis Bate and S. robustics
Smith, but it is easily distinguished from all species hitherto
discovered by the process on the eye-stalks.

Serg. atlaxticus H. Milne-Edw., Bate, p. 389, pis. Ixviii. &,
Ixix.

In my earlier paper I wrote (p. 947) that " I am not convinced
that all the specimens from the localities enumerated (p. 390)
belong to S. atlanticus" and I produced some grounds for that

1903.] IX THE ' CHALLEXOER ' COLLECTION. 59

opinion. The investigation of th© material presei'veil in the
Britisli ]\Tu.seum pi'oved the cori'ectness of my disbelief, but, it
must be admitted, to a degree not supposed.

Of the specimens enumerated by Bate, the following have not
been preserved in the British Museum : " Stat. 42, lat. 35° 58' N.,
long. 70° 35' W., 2425 fathoms," "length 25 mm."— and "On
May 6-1 8th, 1876, in lat. 32° 41' N., long. 36° 6' W. . . one
specimen ... at the surface ; and on the 7th of the same month,
near the Azores .... two other specimens Avere taken at the
sui'face." I can now state with cei-tainty that if the specimen
from Stat. 42 lived near the bottom in that enoi'mous depth, it
did not belong to S. atlanticus ; perhaps it was captured near the
surface, but at all events the locality must be omitted as uncertain.
IMost probably the specimens captured in May 1876 belonged to
8. atlanticus. The specimens from the other stations enumei-ated
by Bate belong to four species, and each station must be mentioned
sepai-ately.

" Korth Atlantic .... Stations 62 and 63, on the passage from
Bermuda to the Azores. Three specimens." In a bottle lal)elled
"Between Bermudas and Azores" I found eight partly mutilated
specimens of S. atlanticus.

" Between Teneriffe and St. Thomas." In a bottle with the
same locality, three specimens of S. atlanticus.

"Station 320, lat. 37° 17' S., long. 53° 52' W., off Monte

Video; depth 600 fathoms." Bate does not directly state the
number of specimens, but he writes " Length 38 mm.," and the
meaning is probably that he had one single specimen. In a bottle
labelled " off Monte Video" I found six small and badly preserved
specimens of >S'. atlanticus M.-Edw., and one large S2^ecimen of
S. arcticus Kj*., and it is decidedly the last-named specimen which
Avas procured from 600 fathoms. As to S. arcticus Kr., the
student is referred to Kroyer's work, to the description and
drawings in various papers of S. Smith, to notes in n}y earlier
paper, and to some i-emarks below, in the description of iS. siviilis,
n. sp., together with figs. 1 a-c on PI. XII.

" Station 159, . . . south of Australia ; depth 2150 fathoms. . . .
Three specimens. . . Length 43 mm." In the collection three
large specimens are present, but they belong to 6'. arcticios Ki*.,
w'hich thus has been proved to be distributed through the deeper
Atlantic, from the southeiii part of Greenland to lat. 38° S., and
to south of Austi'alia.

Bate enumerates still two deep-water stations, viz. Stat. 232,
off Japan, and Stat. 173, off Matuku, Fiji Islands; he examined
one specimen from each of these localities, but the animals belong
to two new species to be described below.

But before giving these descriptions I will sum up the results
of the examination of the specimens referred by Bate to S. atlan-
ticus H. Milne-Edw. ( =*S'.//-mi Kr.). The Copenhagen ]\[useum
possesses some hundreds of adult specimens of S. atlanticus^ taken
at a large number of places in the Atlantic, the Indian Ocean,

60 DE. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

and the western part of the Pacific, and all these were taken
near the surface ; the ' Challenger ' specimens taken near the
surface and referred by Bate to S. atlcmticus really belong to that
species, but his specimens secured at the deep-sea stations 320,
159, 232, and 173 belong to three other species. — I have found it
unnecessary to attempt to point out what pai'ts of Bate's long
description (and which of his drawings) can be applied to jS. atlan-
ticus ; the zoologist is referred to the good desci'iption of Kroyer
together with the additional notes in my earlier paper.

Serg. similis, n. sp. (Plate XI. figs. 6 «-6 d.)

The type of this species is the above-mentioned specimen on
which Bate writes (p. 320) : "Stat. 232, . . . lat. 35° 11' N., long.
139° 28' E., ofi" Japan ; depth 345 fathoms. . . . Length 50 mm."
The species is closely related to S. arcticus Kr. The rostrum
(fig. 6 b) is nearly oblong-triangular, a good deal longer than in S.
arcticus (PI. XII. fig. 1 h) and directed more upwards ; the anterior
margin of the scutum below the rostrum is strongly convex (fig. 6a)
and protruding, while it is nearly vei'tical in S. arcticus (PI. XII.
fig. 1 a). The supra-ocular and hepatic spines are well developed ;
the gastro-hepatic gi'oove distinct. The eyes are nearly as in
/S. arcticus, large, broader than deep, scarcely as long as broad
(fig. 6 c), decidedly shorter than the outer margin of the distal
joint of the stalk, and one-half shorter than its inner maxgin.
The antenn. ped. (fig. 6 c) is about as in *S'. arcticus ; the outer
margin of the basal joint is as long as the same margin of the two
distal joints together ; the second joint is rather slender, its inner
margin almost four times longer than its breadth and scarcely
longer than the third joint, which is slender, about six times
longer than broad. The squama of the antenna is moderately broad
at the end, with the outer spine projecting beyond the terminal
margin. Of the long appendages, mxp.^, trl.^, and trl.' are wanting,
and the remaining thoracic legs are about as in S. arcticus. The
branchise above trl.^ and trl.^ (fig. Qd) present excellent difierences
from those in S. ai-cticus (PI. XII. fig. 1 c) ; br.^ above trl.^ is not
quite half as long as br. above trl."^, and in spite of this not incon-
siderable length it is semi-rudimentary, having only some posterior
branches, nearly all short, and no anterior branches. Of the
branchiee above trl.^, br. is long, four-fifths as long as br. above
trl.'^ ; br.^ is more than half as long as br., very well developed, at
least as long and more than twice as broad as br.^ above trl.'
[In *S'. arcticus (PI. XII. fig. 1 c) — from a specimen obtained in the
most northern part of the Atlantic — br.^ above trl.^ is about three-
fourths as long as br. above trl.^, with well- developed branches
on both sides ; of the branchiae above trl.'*, br. is only a little more
than half as long as br. above trl.^, while br.^ is small, considerably
narrower and shorter than br.^ above trl.''] The ext. br. of urp.
has the apical portion broken off", but it is narrow, and seems to
h% ve been a little more than five times longer than broad, thus
sli ghtly narrower than in S. arcticus, but otherwise of the same

1903.] IN THE ' CHALLENGER ' COLLECTION'. 61

shape aii'l with the marginal spine well developed. — Length
54 mm.

The species is easily distinguished from *S'. arctlcus Kr. hy the
shape of the rostrum, together with the anterioi- margin of the
scutum, and especially hy the branchiae above trl.^ and trl.' ; in
all other features these two species aie closely allied.

Serg. challengeri, n. sp. (Plate XII. figs. 2 a-2 n.)

The type of this species is the specimen on which Bate writes
(p. 390): "Station 173, Julv 24th, 1874; lat. 19^ 9' 35" S.,
long. 179= 41' 50" E. ; off Matuku, Fiji Islands; depth 315
fathoms; bottom, coiul mud. One sjiecimen, male. Dredged.
Length 24 mm." He referred it to *S'. atlanticus. I have the
pleasure of appending the name of the renowned ship to this
Sergestes, which is one of the most interesting species of Crustacea
secured by the expedition. Unfortunately the single specimen is
very mutilated.

The rostrum (fig. 2 b) is luther low, short, obliquely triangular,
turning somewhat upwards; its apex is iicute and very sliglitly
produced. The supra-ocular spine is wanting ; the hepatic spine
is rudimentary ; the gastro-hepatic groove (fig. 2 a) is deep, and
the cervical groove very distinct. The eyes have been broken off,
only the basal part of the stiUks being left. In the antenn. ped.
(fig. 2 c) the outei- margin of the basal joint is a little longer than
that of the two other joints together; the second joint is
moderately robust, its inner margin a little more than three
times longer than the breadth ; the third is slightly more than
two and a half times longer than broad, a little shorter than the
outer margin of the second joint, and only three-fifths as long as
the inner margin of the last-named joint. The squama of the
antenna is distally very broad (fig. 2 c), with the outer spine
scarcely projecting beyond the terminal margin. Mxp.^ and trl.^-
trl.' are wanting ; of trl.'^ the apical part has been lost, but these
legs seem to have been a little longer than mxp.'-, and to be more
slender than in S. arcticus, but otherwise not showing any
difference of importance. Of the branchi;p (fig. 2 d), br. above
trl.'^ and trl.^ are long and narrow ; &r.^ above trl." is as usual a
lamella ; br.^ above trl.' is slightly more than one-third as long
as br., especially with its anterior branches well developed ; of the
branchi?e above trl.', ?)r. is alwut three-fourths as long a,s 6r. above
trl.', while br.^ is proportionately large, nmch longer and broader
than br.' above trl.\ and even more than half as long asb)-. above
trl.' The ext. br. of urp. (fig. 2 1) has the apical portion
wanting, but the branch seems to have been nearly five times
longer than broad, with the marginal spine well developed and
situated as in S. arcticus. — Length 23 mm.

By the shape of the joints of the antenn. perl., the development
of the branchiiP above trl.' and trl.\ and the shape of the ext. br.
of urp., this species is related to ^S*. robustus Smith, .S'. preJiensilis
Bate, and S. kruyeri Bate. But it is sharphj distiiKjuished from

62 DR. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

cdl other species hitherto known hy possessing an enormous multitude
of luminous organs arranged regularly on the lower surface and
near the lower lateral margins of the cephalothorax, on the six
abdominal segments (figs. 2 of, i, h), on the sides of the shield
(fig. 2 a), and on all the appendages preserved (figs. 2 c, d, e,/, g, h)
with exception of the maxillulse and maxillse. The organs are
easily seen ; they look almost similar to the eyes in Arane^e, and
they differ much in size and direction. I have deemed it advisable
to deal with the special arrangement and the structure of these
organs in a separate section of this paper.

Serg. dorsispinalis Bate, p. 394, pi. Ixxii. fig. 1.

Bate does not state directly the number of specimens, but he
had probably only one. The length is " 9 mm." and the locality
"south of Australia, March 1874." An animal with a label of
exactly the same contents was preserved in a microscopical pre-
paration. It is a Mastigopus i-elated to S. arcticits Kr., &'. similis,
n. sp., &c., but it could not be referred to any adult form. In
the preparation the abdomen is seen from the side, and the
cephalothorax essentially from below; the spine on the scutum
on which Bate writes "just anterior to the [cervical] suture, in
the median dorsal line, is a small, anteriorly directed tooth," is
in reality the gastro-hepatic spine, which besides has been drawn
in a position too remote from the front. Bate's figure is not
correct in some other respects : in the antenn. ped. the second
joint is about as long as the third, and both together somewhat
shorter than the first ; the eyes reach to the end of the first joint ;
the squama is longer than in the figure, reaching almost beyond
the second joint of the antenn. ped. and distahy narrow ; the ext.
br. of urp. is a little longer and conspicuously more narrow than
in the figure, while its marginal tooth is indicated correctly.

Berg, laterodentatus Bate, p. 395 (no figure).

Bate has examined one specimen, measuring 8 mm. in length,
and captured " south of Australia, March 1874." The type could
not be found in the Museum. According to the description it is
a Mastigopus belonging to the arcticus-grouTp, probably a younger
stage of the same species as that to which the preceding larva,
>S'. dorsispinalis Bate, belongs.

Serg. nasidentatus Bate, p. 398, pi. Ixxii. fig. 2.

Bate does not state directly the number of specimens, and
presumably he had only one ; the length was 10 mm. and the
locality the Pacific Ocean, between Valparaiso and Juan Fernandez.
The type does not exist in the Museum. The species is a
Mastigopus ; in my earlier paper I had already referred it to the
same group as S. arcticios, and nothing further can now be added.

Serg. diapontius Bate, p. 399, pi. Ixxii. fig. 3.

Bate does not state the number of specimens, but presumably

19*^3.] IX THE 'CIIALLEXGER' COLLECTION. 63

he had only one. The length is 18 mm. and the locality the
Atlantic Ocean, AprO 7, 1876. In a preparation bearing the
name of the species, and besides " Surface, 7 April, 1876
Atlantic," one specimen is preserved : it must be Bate's type'
but it measures only 16-5 mm. in length. In Bate's figure the
armature on the dorsal line of the abdomen is not correct • on
the third segment a spine has either been broken off or is rudi-
mentary, the base being distinct ; the spine on the fourth segment
is only half as long as that on the fifth, shorter than in the figure,
and directed obliquely backwards ; on the sixth segment a%ery
short sphie is visible. The basal part of the rostrum is somewhat
ascending, the distal spiniform and horizontally porrected. The
ext. br. of urp. has the outer margin hairy in about | of its
length, and the marginal tooth is very small. The distal" part of
mxp.' is very incorrectly drawn by Bate in his fig. 3 z ; it is four-
jointed ; the third joint is short, much shorter^ than the fourth,
and both together about as long as the second ; the first of these
joints terminates below in two strong setiform spines, both some-
what longer than the second joint, which terminates in two spines
of the same length as the preceding jDair ; the fourth joint
terminates also m two spines, somewhat, but not much, shorter
than the four just mentioned. (I cannot understand how Bate
drew his misleading figure ; it must prevent eveiy student of
his Report from arriving at a correct judgment.) In the British
Museum I have drawn tolerably accurate sketches of the rostrum
the ext. br. of m^p. and the distal part of mxp.^ and I have
compared them with a few specimens in the Copenhagen Museum
previously determined and shortly described by me as the 2Iasti-
gopus of S. penerinkii Bate, H. J. H. : I found the agreement to
be so close, that I must consider S. diapontms Bate and S pene-
rinkii Bate (the type of the latter form unfortunately is not present
in the Museum) as two Mastigopiis-^\.&g%^ of the same species ;
the type of S. penerinUi Bate measured only 8 mm. in leno-th
and is a young Mastlgopus, while ^. diapontius Bate is the la?oe
lai'va. ^

Bate describes S. diapontms on p. 399, S. loenerinkii on p. 418-
and the name .S'. diapontius must therefore be accepted for the
species. In my earlier paper I described the black-eyed adult
form as S. penerinkii Bate, H. J. H., but I think that it must
now be necessary to adopt the name S. diapontius, not only for
the Mastigo^nis, but also for the mature form, which therefore in
the future must be named ,S'. diapontius Bate, H. J. H.

Serg. armatus Kroyer, Bate, p. 401, pi. Ixxiii. fig. 1.

io-^?o^^T?'''''^''t*^'''^^ localities: one of them Ts " September
12, 18/5, between Japan and Honolulu. South Pacific Ocean"-
the second IS - Station 256, July 21, 1875; .... north of the
Sandwich Islands ; depth 2950 fathoms." But in the collection
I found a statement aberrant from both, viz. : " Surface. Japan
to Honolulu, July 1875. Type." This bottle contained Iwo

64 DR. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

small specimens, but neither of them can be the type for Bate's
figure. Both specimens belong to Group II. in my earlier j)aper,
but neither of them belongs to S. armatus, both having on the
rostrum a well-developed sub-basal dorsal spine, which is absent
in S. armatus Ej". and in Bate's figure. I have been unable to
refer the specimens, which measure about 8*5 and 11 mm., to any
species known to me, and I thought it tiseless to describe and
figure them. — One specimen from Port Jackson, Australia, the
third locality in Bate's report, measures at most 6"2 mm. without
the rostrum, which has been broken at the middle, but possesses
a very fine svib-basal spine. It is so small and so badly preserved
that a reference to any species has been impossible.

Finally I found a specimen from "Sidney," determined by Bate
as S. armatus, but not mentioned in his woi-k. It is only as long
as the preceding specimen and impossible to determine.

The result is that I have perhaps not seen the specimen figured
by Bate, which may belong to aS*. armatus Kr., and, according to
the explanation of the plate, measured about 10 mm. in length,
while his specimens of this length examined by me disagree with
his figure by possessing a sub-basal upper spine on the rostrum.

Serg. edwardsii Kroyer, Bate, p. 403, pi. Ixxiii. fig. 2.

Bate enumei-ates three localities. The first is " North Atlantic,
April 1873 " : in the Museum I found a specimen labelled
"14 April, 73, oS" Africa, surface," which most probably is that
indicated in the text, and it belongs to 8. edioardsii Kr. The
second locality is "Pacific Ocean, surface, September 1875": in
the collection a specimen bearing the same inscription is>S'. edwardsii
Kr. From the third of Bate's localities. Cape Yerde Islands, I
found no specimen, but a specimen without locality and signed
" type " is an adult specimen of S. edivardsii Kr. (That Bate's
statement " Greenland [Kroyer) " is wrong here, and in almost all
other places, has already been pointed out both by Ortmann and
myself.) — Furthermore, I found two small specimens of S. ocidatus
Kr., the Mastigopus of S. edioardsii Kr., which had been deter-
mined by Bate as S. edioardsii and labelled " Aug. 23, 1873,
lat. 2° 25' N., long. 20° 1' W., 100 fathoms," but these specimens
are not mentioned in Bate's Report.-

Bate's description of the characters of S. edioai'dsii Kr. is in-
complete ; the reader is i-eferred to my earlier paper.

Serg. rinkii Kroyer, Bate, p. 404, pi. Ixxiii. fig. 3.

Bate mentions two localities : " New Hebrides, August 23,
1874," and "South Pacific, 1875." From the first of these
localities the anterior half of a specimen was present. Further-
more, I found two specimens labelled " Oct. 19, 1875, S. Pacific,
drawn," and one and a half specimen labelled "Oct. 18, 1875,
surface" : both these localities are in all probability identical with
the second one in Bate's Repoi't. It may be very possible that all
these specimens belong to S. rinkii Kr., which is the Mastigopus

1903.] IX THE 'challenger' collection. 65

of S. arcticiis Kv., but liaving in Lniidon no matcrinl from the
North Atlantic of S. rinkil Kr., for direct comparison of niinnte
details, and some other species allied to S. arcticas being known
from the Indian Ocean and the Pacific, I could not determine the
larvaj enumerated with absolute certainty.

One specimen from Cape York, detei-mined as &'. rinkii, but
not mentioned in the Re^iort, is 6'. comic alum Kjr., in a stage a
little younger than that figured by Bate as the last-named specias.

One specimen, labelled " 25. 8. 73," determined as S. rinkii,
but not mentioned in the Repoi't, measures nearly 7 mm. ; it
belongs to S. ijenerinkii Jiate, the young larva of S. diaponilas
Bate, H. J. H.

Seeg. oculatus Kroyer, Bate, p. 406, pi. Ixxiv. fig. 1.

Bate enumerates six localities. From Stat. 106 three specimens
were obtained, from Stat. 257 one specimen, from Stat. 103 one
specimen, finally from " September 12, 1875, South Pacific," one
specimen : all six specimens correctly referred to the Ki'oyeriau
species, which is the Masticjopiis of S. eclwardsii Kr. (compai-e my
earlier paper). From the two remaining localities, viz. " North
Pacific, near the Sandwich Islands, August 21, 1875," and
"August 27, 1873 . . . ofi" St. Paul's Rock," no specimens could
be discovered.

Serg. ovatoculus Bate, p. 408, pi. Ixxiv. fig. 2.

Bate gives the locality " The North Atlantic Ocean." In the
collection I found three specimens with the label '"14 June, 1873,"
which agrees with my quotation from the text. These three
specimens ai'e identical with *S'. avcylo'ps Kr., which is the
Mastigojyihs of S. atlanticus IT. Milne-Edw.

Serg. parvidens Bate, p. 409, pi. Ixxiv. fig. 3.

Bate has the . following localities: " The tropical part of the
Atlantic ; Pacific Ocean, north of the Sandwich Islands ; ofi"
Sydney and Wellington, Australia." Just beloAv he writes:
" Specimens of this species or vaiiety were taken during the
passage from Teneriffe to St. Thomas" ; and in the collection one
specimen from the last-named locality is present : it belongs to
*S'. vigilax Stimpson, the Mastigojnts of S. vigilax Stimps., H. J. H. ;
it agrees with Bate's description and figure, and it had already
been pointed out by Ortmann and myself that *S'. 2>(i'''"^'idevs Bate
belonged to *S'. vigilax Stimps. Examples from the other localities
mentioned by Bate could not be found. — From " Cape York" two
specimens determined as 6'. 2>(i^"videns were present : one of them
is a young S'. ocidatus Kr. (the Mastigopus of «S'. edirardsvi Ki-.);
the other is allied to 6'. incertus H. J. H., but is so badly preservett
that a determination was impossible. In a preparation a small
specimen from the " China Sea," determined as *,S'. parvidens, was
preserved ; it seems to be S. ocidatas Kr., but it is in a very bad
condition.

Proc. Zool. Soc— 1903. A^ol. I. No. V. 5

66 DR. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

Serg. corniculum Ki'oyer, Bate, p. 410, pi. Ixxv. fig. 1,

Bate writes : " Cape York ; north of New Guinea ; ISTorth-west
Pacific." In the collection I fovmd one specimen from Cape York
labelled " type," furthermore, one and a half specimen from the
same locality, all correctly I'eferred to the Kroyerian species. In
a preparation I found a specimen determined as S. corniculuin
from "ISr.W. Pacific"; it measures 10-4 mm. in length, but
according to the shape of the ext. br. of urp., the eyes, &c., it is
not that species but S. ancylops Kr., the Mastigopus of S. atlanticics.
From the thii-d locality, " north of New Guinea," no specimen
could be found.

Serg. ancylops Kroyer, Bate, p. 413, pi. Ixxv. fig. 2.

Bate has two localities : " New Hebrides ; Pacific, July 20,
1875." I found one specimen from "New Hebrides" correctly
referred to the Kroyerian form. — In three other tubes, specimens
not mentioned in Bate's text were present ; they belonged to
three other species, but are mutilated and could not be determined.

Serg. longirostris Bate, p. 415, pi. Ixxv. fig. 3.

Bate writes : " Mid Atlantic, April 1 876," and according to
the following line he had one specimen. But in the collection I
found three tubes, each containing one specimen, all determined
as S. lojigirostris, and two of them from " N. Atlantic," while the
third had no locality ; all three specimens are the young Mastigopits
of S. corniculum Kr., H. J. H. The specimen without locality
seems to be the type for Bate's figure of S. longirostris, but it
measures scarcely 3 mm. in length, the rostrum not included, and
in Bate's figure the eye-stalks are too long, the eyes too small, and
the third joint of the antenn. ped. too short (it is in reality as
long as the second).

Serg. junceus Bate, p. 416, pi. Ixxvi. fig. 1.

According to the text Bate has seen one small specimen from
the "South Pacific Ocean." It could not be detected in the
collection. But in my earlier paper I pointed out that it must
be the young Mastigopus of S. tenuiremis Elr., H. J. H.

Serg. longispinus Bate, p. 417, pi. Ixxvi. fig. 2.

Bate has two localities: " Station 106 .. . Mid Atlantic" , . .
and " Station 354 .... North Atlantic Ocean." The collection
contained two specimens referred to this species and labelled
" 23 Aug. 1873, 70 meters "; in the text we find "Station 106,
August 25, 1873 ; lat. 1° 47' N., long. 24° 26' W. ; Mid Atlantic
Ocean " • but in spite of the small difierence as to the date
(probably originating from a missci'ipt) I am sui'e that one of
the two specimens was captured at Stat. 106. The other
specimen is probably from Stat. 354. Bate wi'ites : " the specimen
from the ti'opical part of the Atlantic " and " that from the North
Atlantic," which indicates that he had only one specimen from

1903.] IX THE 'CUALLKXCER' COLLECTION. 67

eacli of the two stations mentioned ; and I suppose, therefore, that
both specimens have latei' on been put together in the sam« tube.
Both specimens, which are in a very bad condition, belong correctly
to S'. longispinus Bate (the direction of some of the dorsal spines
on the abdomen is very characteristic), and this form is in reality
the ^fast^gopus of S. conmtas Kr. (compai-e my earlier })aper,
p. 953).

Serg. pexerixkii Bate, p. 418, pi. Ixxvi. fig. 3.

Bate records tlie length to be 8 nun. and the locality " North
Atlantic Ocean." No specimen referred to this species could be
found in the collection. But al)Ove I have mentioned a specimen
of this species referi'ed by Bate to 6'. rinhii, and it may perhaps
1)6 that desci-ibed by him as S. penerinklL He writes (p. 419)
that the last-named " species bears a strong resemblance to
/S'ergres^es r«iA;« Kroyer, but differs . . .". The specimen in question,
referred by him to S. 7-inkil, measures nearly 7 mm. in leng-thand
was captui'ed " 25. 8. 73," that is to sivy. Mid Atlantic. — In my
earlier paper I have described the Masiigopits, and besides the
mature form as *S'. peneruikii Bate, H. J. H. ; but above it is
pointed out that this name must be cancelled as a synonym, and
the species be named aS'. dlapontias Bate, H. J. H.

Serg. fermerixkii Bate, p. 419, pi. IxxAd. fig. 4.

Bate has examined one specimen, captured in the " Paciiic
Ocean, lat. 24° S., long. 148° W.," and measuring 5 mm. The
specimen is not to be found in the Museum collection. According
to the figui'e it is a very young Mastigopus belonging either to
»S'. diapontius Bate, H. J. H., or to a closely-allied species of the
same group.

Serg. loxgicollus Bate, p. 421, pi. Ixxvii. fig. 1.

Bate enumerates two localities. The fii-st of them is " Soutli
Atlantic Ocean, October 5, 1873 ; near Station 131 ; lat. 29° 35'
S., long. 28° 9' W." In a preparation labelled ''5. 10. 73" I
found the lai'ge specimen diuwn by Bate ; in a tube labelled
"surface, at night, 5 Oct. 1873, South Atlantic," a small specimen
was presei'ved. Thus both specimens are fi'om the fii'st locality
in the text, and both belong to *S'. lo/igicoUas Bate, which is the
Mastigopus of .S'. tenairemis Kriwei', H. J. H. In the large
specimen di'iiwn by Bate the eyes present a thick yellowish Inyer
around the black central part ; the dorsal spine on the sixth
abdominal segment is exceedingly small.

The second locality in Bate's text is " Station 295. November 5,
1875 ; lat. 38° 7' S., long. 94° 4' W. ; South Pacific Ocean ; depth
1500 fathoms; . . . Taken at night." In the collection ti specimen
is preserved labelled " 5. Nov. 75, night, S. Pacific, surface."
The determination is correct ; and the S2:)ecimen, which measures
9"2 mm. in length, is evidently that mentioned by. Bat<? ; but
the label proves that it was taken at the surface, wheiefore

5*

68 DR. H. J. HANSEN ON CRUSTACEANS [Jail. 20^

the depth recorded in the text is roost misleading. — In a third
tube two specimens labelled 18/10 75 were preserved ; they had
been correctly referred by Bate to his >S'. longicollus, but are not
mentioned in his text.

A preparation contained a specimen of the same species,
labelled " Sergestes tenuiremis, South Atlantic, 6. 10. 73." It is
not mentioned by Bate, who even writes (p. 421) that " No speci-
men in the ' Challenger ' collection corresponds precisely with the
description and figure given by Ki'oyer" of S. temoiremis. On
the other hand, ^S*. tenuiremis Kr. and S. longicollus Bate are, as
already stated, the same species.

Serg. PRiECOLLUS Bate, p. 423, pi. Ixxvii. fig. 2.

Bate has examined one specimen, measuring 25 mm., from the
" IsTorth Pacific Ocean." The specimen could not be found.
The shape of the eyes in Bate's figure shows that it has been a
large Mastigo'jnijS. In my earlier paper I wi-ite (p. 958) : " is at
least rather nearly related to S. corniculum Kr., from which it seems
to difi"er by a somewhat different shape of the ext. br. of urp.,
and by having the fifth abdominal segment ' dorsally produced to
a point.' "

Serg. semiarmis Bate, p. 423, pi. Ixvii. fig. 1.

Bate has two localities: "West Pacific Ocean "and "Station
354, May 6, 1876 ; .... Mid North Atlantic." After the
description he writes : " Observations. — A specimen (pi. Ixvii.
fig. 2) very similar to the type was taken in the Atlantic . . . " ;
and then he describes the specimen. I must suppose that he
considered the specimen from the "West Pacific Ocean" to be
the type of his S. semiarmis, and that the specimen described
separately (p. 425) is that from ". . . May 6, 1876. . . ." Neither
of the specimens could be detected in the Museum ; both are
larvse, but a reference is impossible. In the collection I found
one specimen determined as S. semiarmis and labelled "13 April
1876, Atlantic, off" coast of Africa, surface"; it is a Mastigopus
of S. diapontius Bate, H. J. H.

Serg. l^eviventralis Bate, p. 425, pi. Ixvii. fig. 3.

Bate had probably only one specimen, 7 mm. in length, from
" North of New Guinea " ; it could not be found in the Museum.
It is a young Mastigopus belonging to a species closely related to
jS. arcticus Kr.

Serg. spiniventralis Bate, p. 426, pi. Ixvii. figs. 5 & 6.

Bate has only one locality : " North Pacific Ocean " ; the
animal described measured only 3" 5 mm. in length, and some of
its parts are shown in figs. 5 a, 5 Z, and 5 v, but fig. 5 v, repre-
senting the ext. br. of urp., does not correspond at all with the
description in the text (p. 427). The animal is not to be found
in the collection ; it is a small Mastigopus related to S. vigilax
Stimps. and allied species. Bate's fig. 6 a represents the head

1903.] IX THE 'CHALLEXGER' COLLECTION. 69

of " Sergestes sjjiniventrcdis var." from the " West Pacific " ; it
has been suggested that the animal is allied to — or identical
with — the Mastigopus of S. vigilax Stimps., H. J. H., but the
specimen could not be found. The collection contained one
specimen determined as ^S'. spiniventralis and labelled " Sidney to
Wellington, 17. 6. 74"; it measures about 7'5 mm. in length,
and the naked part of the outer margin of the ext. br. of urp. is
slightly more than one- fourth of its length. The specimen is in
all probability a young Alastigojyits of *S'. vigilax Stimps., H. J. H.

Serg. trofuxdus Bate, p. 428 (no figure). (Plate XI. figs. 3 a,
Zh.)

Bate has referred two specimens to this sjoecies. He describes
each specimen separately: the first of them, from Stat, 137,
belongs to Fetalidiiom, perhaps to P. foUaceum, and has been
dealt with above. The other specimen, from " Stat. 300,
December 17, 1875; lat. 33° 42' S., long. 78° 18' W. ; west of
Valparaiso ; depth 1375 fathoms ; . . . . Ti-awled," is a real
Sergestes, related to 6'. inous Faxon, but differing in the shape of
the rostrum. Having removed the first-mentioned specimen
from the genus Sergestes, I should think it justifiable to apply
the name S. profunchis Bate to the last-named specimen, instead
of proposing a new name. The animal is quite membranous,
and is crimson everywhere — a fact proving that it lives in the
depth of the sea, and that its colour has been dm-able to the
highest and most unusual degree. The posterior part of the
abdomen is Avanting, and the animal is on the whole so mutilated
and flabby, that I would have omitted it if it had not been
described by Bate ; but for that reason I have thought it necessary
to add some notes with two figures (PL XI. figs. 3 a, 3 h). It
agrees with ^S*. inous Fax. as to the membranous quality of the
skin and the posterior branchite, but differs from it in the shape
of the rostrum, which is of moderate length, with the upper
mar-gin somewhat, and the lower margin paitly, sti-ongly convex,
and distally it is pi'oduced in a moderately shoi't spine (PL XI.
fig. 3 h). (Unfortunately I have not seen any specimen of the
gigantic species S. inous Fax., and can therefore not decide whether
the specimen described by Faxon had the rostrum uninjured.)
Supra-ocular and hepatic spines are wanting. The eyes (PL XI.
fig. 3 a) are black, compaiutively small, somewhat shoi'ter than
broad, not half as long as the eye-stalks, and not broader than the
distal end of the stalks. In the antenn. ped. the outei' margin
of the first joint is almost as long as that of the two distal joints
together, therefoi-e somewhat shorter than their inner mai'gin ;
the thix'd joint is a little shorter than the inner margin of the
second, and seems to be about thi'ee and a half times longer than
deep. The squama does not reach the end of tlie antenn. ped.,
and the outer distal spine is well developed. Accoi'ding to Bate
the part preserved measures 18 mm., and the probable length of
the entire animal is 24 mm.

70 DE. H. J. HANSEN ON CRUSTACEANS [Jail. 20,

Serg. ventridentatus Bate, p. 431 (no figure).

Bate gives the locality " north of the Sandwich Islands " and
the length " 7 mm." A specimen labelled in accordance with the
text is preserved in balsam ; it is a young S. oculatus Kr., the
Mastigopus of S. edwardsii Kr.

Serg. utrinquedens Bate, p. 433 (no figure).

Bate gives the locality " North Pacific Ocean," and the length
" 3"5 mm." No specimen could be found in the collection. Bate's
specimen is a very young Mastiyopits ; in my earlier paper I
have placed it near S. comiculum Kr., but a final interpretation
is impossible to me.

Serg. DissmiLis Bate, p. 437 (no figure).

In the collection one specimen, from Cape Verde Islands, is
preserved ; it is certainly the type described by Bate. In my
earlier paper I had determined it as one of the larval stages of
aS'. arctic'us Kr., but this is not correct. The rostrum is slightly
more than one-third the length of the eye-stalks, its basal part
somewhat ascending, with a very small spine on the upper angle,
its distal part is a slender horizontal spine. The hepatic spine is
short. The eyes are only a little higher than the distal end of
the eye-stalks, and these increase gradually in thickness from the
base outwards. In the antenn. ped. the distal half of the first
joint has the margins nearl}^ parallel, and the two other joints ai-e
a little thicker than in the cori-esponding Mastigopus of S, arcticus.
Fourth to sixth abdominal segments each with a very small dorsal
spine directed backwards ; the spine on the sixth segment is the
longest. The ext. br. of urp. is almost, but not quite, five times
longer than broad. All these characters agree with those found
in a Mastigojjus of ^S'. mediterraiieus H. J. H. preserved in the
Copenhagen Museum ; and sketches drawn in London of the shape
of the distal part of the squama and the telson agi^ee also with
the last-named form. I must therefore consider S. dissimilis Bate
as identical with *S'. mediterraneus H. J. H. The result is that
the last name must be withdrawn as a synonym, and the species,
of which the sut-adult stage has been described in my earlier
paper, mvist be called S. dissimilis Bate, H. J. H.

II. Some Residts of the Investigation.

In my earlier paper on Sergestes I have paid as much attention
as possible to the animals described by Bate ; in nearly all cases
I was able to state whether the specimen was an adult form or a
Mastigoptis, and several of the species were interpreted. After
the study of the material preserved in the British Museum, I have
now been able to confirm most of my earlier statements, to
interpret an additional number of the specimens mentioned by
Bate, and to correct two faults committed by myself. I had
ei'foneously referred S. dissimilis Bate to S.^ arcticus Kr., instead
of identifying it with S. mediterraneus H. J, H. (see above).

1903.] IX TUE 'challexger' collectiox. 71

Furthermore, I had divided a number of species, enumerate! on
p. 949 as Group I. A. h. fy, into two sections, according to
difierence in the thickness of the distal joints of the antenn. ped.;
but a study of the types of S. 2>^'6hensilis Bate and S. kroyeri
Bate showed that Bate's drawings of the antennul;ie of these
species ai-e incoirect and misleading, wherefore my ari'angement
of them wtvs wi-ong.

It may be useful to put together the alterations and additions
which may now be accepted in the Consj^ectus on pp. 949-51 in my
eai'lier paper. In Group I. the following particulais must be
added or altered : — To *S'. atlanticus H. M.-Edw. belongs only a
part of S. atlanticus sens. Bate, besides the foi-m i-efeixed by Bate
to S. ancylops Kr. From S. arcticus Kr., aS'. dissimilis Bate must
be removed, and the latter species is to be established separately
with 6'. mediterraneus H. J. H. as a synonym ; f ui-thermore, some
of the specimens referred by Bate to S. atlanticus belong to
S. arcticus. Near S. arcticus Kr., must be insei-ted S. similis
H. J. H., established on one of Bate's specimens of S. atlanticus.
S. prehensilis Bate and S'. kroyeri Bate must be removed fi-om
their place and inserted above near S. japonicus Bate, together
with aS. 2^rofuvdus Bate, in its new restriction, and S. challengeri
H. J. H., established on one of the specimens referred by Bate to
S. atlanticus. S. longirostris Bate must be withdi'awn as being
a Mastigopus to S. cornicidum EJr., H. J. H., and S. corniculum
sens. Bate is the same species. As uncertain remain : S. dorsi-
spinalis Bate, ^S*. laterodentatus Bate, S. nasidentatus Bate, S. Icevi-
ventralis Bate, *S'. rinkii Bate, ? KJr., all larval forms belonging to
species related to S. arcticus Kj.\, or perhaps partly belonging
to S. arcticus itself ; furthermore, the larvas S. prmcollus Bate,
S. utrinquedens Bate — both at least rather closely related to
S. cormculu/m Kr., H. J. H., — and S. semiarmis Bate.

In Group II. there is hardly anything to alter, but some additions
to make. To S. edwardsii Kr. belongs *S'. edwardsii Kr., Bate,
S. oculatus Kr., Bate, S. intermedius Bate, and S. ventridentatus
Bate. S. jjenerinkii Bate must be cancelled as synonymous with
aS'. diapontius Bate ; and the adult form described as »S'. penerinkii
in my eai-lier paper must be named ;S'. diapontius Bate, H. J. H.
The rest of Group II. remains unaltered. S.fermerinkii Bate,
S. spiniventralis Bate, and the species referred by Bate to
S. armatus Kr. coidd not be interpi'eted.

Besides, the investigation has yielded some results on the
bathymetrical and geographical distribution of some of the species.
It has been proved that the large specimens (exceeding 30 mm.
in length) referred by Bate to S. atlanticus Kr, are deep-sea
forms belonging to other species : >S'. atlanticus is common near
the surface; according to Ortmann it has been taken in the
intermediate net from 700-500 m., but it does not live in the
greater depths of the sea. S. arcticics Kr. has a very wide
geographical range, through the deeper to very deep tracts of the
Ocean (see above); S. japonicV'S Bate has been captured in
the northern part of the Atlantic and the northern pai-t of the

72 DR. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

Pacific. ^S'. atlanticus H. Milne-Edw., S. edwardsii Kr., S. vigilax
Stimps., H. J. H., S. tenuiremis Kr., H. J. H., and .S'. corniculum
Kr., H. J. H., have been proved to be distributed through the
tropical and subtropical parts of the Oceans almost around
the globe, viz., from lat. 23° N., lat. 32° N., or even lat. 42° N.
in the Atlantic, throughout the Indian Ocean to Australia, New
Hebrides, and " South Pacific."

III. Limiinous Organs in Sergestes challengeri, n. sp.
(Plate XII.)

The luminous organs briefly mentioned above are generally
easy to discover ; each resembles a very convex, vitreous, faintly
yellowish lens, which is circular and sharply defined. They differ
much in size, some of them being very small and many pro-
portionately large. It may be advisable, first, to give a con-
spectvis of the oi'gans observed on the single and tmfoi'tunately
very mutilated specimen, next to add some remarks on their
distribution, then to describe their structui-e, finally to compare
them with luminous organs in other Crustacea.

Conspectus of the Organs observed.

Organs.

On each side of the scutum a row with four organs. 8

On the clypeus one oi'gan 1

On the lower side of the head one unpaired organ

and one near the lateral mai'gin 3

On the lower surface of the thorax and on its inf ero-

latei'al anargins 26

On the lower side of the third joint of the antenn.

ped. one organ 2

On the lower side of each squama four oi-gans 8

On the mandible and its palpus two organs 4

On the first maxilliped two organs 4

Oil the second maxilliped five organs 10

On the first trunk-leg three organs 6

On the lower surface and on the lateral wall of the

first abdominal segment , 6

Do., do. of the second abdominal segment 6

Do., do. of the third abdominal segment 4

Do., do. of the fourth abdominal segment 3

On the latei'al wall of the fifth abdominal segment. 2

Along the median line of the sixth abdominal segment. 6
On the outer side of the basal joint of each of the

pleopods one organ 10

On the peduncle and on the inner branch of each

uropod two organs 4

On the lower side of the outer branch of each ixropod

two organs 4

Total number of organs ... 117

1903.] IN THE ' CnALLEN'GER ' COLLECTION. 73

The eye-stalks, the moxillipeds, and the four posterioi' pairs of
the trunk-legs have been broken ofi". I am convinced that at
least most, and perhaps all, these appendages possess some oi-gans,
and the lowest number the species possesses must therefore be
about 150 !

On the distribution and direction of these organs the following
remarks may be offered. The four orgjins on each side of the
scutum are arranged i-ather close in a longitudinal i"ow situated
on the ridge bordering the branchial cavity above (fig. 2 a). The
organ on the clypeus is lai-ge : one oi'gan is situated on the
segment bearing the antennulre, and one on the lower surface of
the head near the lateral margin a little in front of the mandible ;
these four oi'gans look essentially dow^lwaI■ds. The arrangement
of the organs on the three posterior thoracic segments can be seen
in figs. 2 i and 2 d. It is observed that two large organs are
placed at the lower margin of br. above trl.^ and of br.^ above
trl.'' These organs look outwards and downwards, ard the part
containing the glandular mass etc. behind the lens is somewhat
protruding, which produces an aspect as if these organs had been
inserted on the end of a kind of rather thick, short stalk. The
remaining oi-gans on the segments mentioned are found on the
lower surface (fig. 2 i) ; those placed at the inner angle of the
legs are small oi- very small, while some in the median line are
large. In fig. 2 i fifteen thoracic oi'gans have been drawn ; the
remaining eleven thoracic organs ai-e situated on the antei-ior
segments and arranged in a rather similar way. The four organs
on the lower surface of each squama of the antenna are arranged
for some distance in a row ; two of them are seen through the
sfjuama in fig. 2 c. The mandible (fig. 2 e) has one organ below
at the antero-inferior margin near the insei'tion of the palp ;
another oi'gan is seen on the lower surface of the first joint of the
palp near its distal end. The first maxilliped (fig. 2/) has on
the upper side one organ just at the origin of the exopod, and one
on the lower side of the following joint of the endopod ; the fii'st-
nained organ looks forwards and a little inwards (fig. 2 g), and
the upper mai-gin above it is produced nearly as a lamella, over-
lapping the major part of the lens wdien seen from above (figs. 2 (j
<fc 2 n). Of the oi-gans on the outei' — in the natural position of
the appendage in reality the lower — surface of mxp." (fig. 2 k),
tha.t at the base of the third joint is very small and the others
large ; the oi-gan on the first joint looks essentially downwai'ds,
and is " stalked," as the above-described organs near the lower
margin of the posterior branchiae. Of the three organs on trl.\
two are placed on the inner side of the long fourth joint, one near
the base and the other near the distal end ; the third organ is
situated on the anterior side near its end. Each of the five
anterior abdominal segments has a lai'ge organ on the lower pai-t
of the antei'ior margin of the lateral plate (figs. 2d & 2 1) ; it
looks forwards anrl somewhat downwai-ds, besides sometimes a
little outwards. Each of the two antei'ioi- segments has besides

74 DR. H, J. HANSEN ON CRUSTACEANS [Jan. 20,

on the lower surface (fig. 2 i) four organs, one of which is large
and two of the others very small ; on the third and fourth
segments these organs are gradually reduced in number, and none
of them are found on the fifth. The sixth abdominal segment
has a median row of six organs, which are seen in fig. 2 k with
the exception of the first, this being hidden by the lateral wings
of the preceding segment. The very short basal joint of the
peduncle of each of the pleopods has on the outer side a small
organ, looking outwards and at least sometimes a little down-
wards. Each lu'opod has one organ on the pedvmcle, situated on
its inner surface near the lower margin and close to the base,
besides one organ on the interior (lower) surface of the inner
branch near its base ; finally, two organs on the interior (lower)
surface of the outer branch, one of them near the middle and the
other on the distal nari'owing part.

From the preceding description it is seen that most of the organs
look downwards, a smaller number somewhat outwards or forwards,
rather few almost totally outwards, and none upwaixls, with
the exception of one on the first maxilliped. With the exception
of the few lateral organs on the scutum, all the others are found
on or near the lower surface of the body and on the appendages.

The structure of the organs is very intei'esting, and very
difierent from all hitherto known in any invertebrate animal. I
have examined more closely three organs, viz., that situated on
the third joint of the first maxilliped, one of the "stalked"
organs near br. above trl.^, and one from the antero- inferior
margin of the fourth abdominal segment. The two last-named
organs have been cut off, most of the adhering tissue removed
by dissection, and the organs examined with moderately high
magnifying power. I ha,ve found no difi"erence of any importance
between the three organs from such distant parts of the animal's
body.

The organ taken as type is that from the infero-lateral margin
of the thorax ; it has been drawn (fig. 2 ni) in optical vei'tical
section. The skin forms a chitinous, large, and very thick biconvex
lens (a), which is vitreous and a little yellowish ; the major
central part is covered by a rather thin limpid layer («^), but this
layer I could not perceive on the two other organs examined.
The lens is circular in outline ; its diameter is about two-thirds
as long as that of the inner portion of the organ at its thickest
part. The inner side of the biconvex lens is covered by a large
and rather thick concavo-convex lens (6), which is somewhat
thinner than the outer lens, but with the diameter a little longer
than that of the same ; the lateral margin of the inner lens is
oblique, very broad, touching the external chitin around the
outer lens. This inner lens, which consists of two layers, is
homogeneous, vitreous, and slightly greyish ; but the difference
between the colour of the outer and the inner lens is in the figure
purposely a little more strongly marked than in nature. These
two lenses remind one of optical instruments in which the lens is

1903.] IX THE ' ClIALLEXGEK ' COI.LECTIOX, 75

composed of crown-glass and flint-glass. Behind the inner lens
is found a thick layer of glandular cells {d), A\hich are light
greyish, very lai'ge, and most of them elongate, I'adiating towards
the centre of the outer lens. The diameter of this layer is some-
what larger tlian that of the inner lens ; and when the luminous
organs are examined in their natural position with a strong
pocket-lens, this layer can often be seen through the skin as a
whitish ring around the lens. Between the layer of large cells
and the inner lens a thin layei- (c) seems to exist, but its quality
could not be made out, and I do not venture to pi'ojoose any
hypothetical explanation. Behind the glandular layer is seen
another (e), Avhich is yellowish, with numei-ous ti-ansverse fine
stiipes, and without ti-ixce of cellular sti-ucture ; it is i-ather thick
in the middle, but thin around the sides of the glandiiiar layer.
The internal surface of the organ seems to be covei'ed l)y a thin
layer (/') of connectiv^e tissue. That the posterior layer with the
ti-ansvei'se stiipes is — as in the EuphaiisiicUe — a I'eflector can be
taken for granted. But it is impossible to decide wdiether the
light is pioduced by the glandular layer or by the inner lens.
AVhether the thin layer enveloping the whole organ is pigmented
or not cannot be seen on this old material, which has been
pi'esei'ved twenty-eight yeai-s in spiiit ; that the oi-gans ai'o
immovable scarcely needs to be stated. Future investigation of
living animals and of sections of fresh inateiial must elucidate
whether the organs are especially innervated or not, and, besides,
fill up the other gaps in the interpi'etation of the function and
structui-e of the layers in these compound structures.

A bi-ief compai-ison of the luminous organs in Sergestes
challengeri with those in other Crustacea is not without interest.
Of animals belonging to that class, luminosity has been observed
in some Copepods, a few Ostra.cods, Euphausiidpe, and one
macrurous Decapod. Giesbi'echt has publislied a thorough and
interesting study : " Ueber das Leuchten der pelagischen
Copepoden und das thierische Leuchten im Allgemeinen" (Mitth.
zool. Stat. Neapel, 11 B., 1895, pp. 648-689). "He has examined
a small nuinlier of jielagic Copepoda — necessaiily I'estrieting
himself to all the luminous forms which he could jjrocure in a
living state in the Gulf of Ka.ples — showing that these animals
possess a number of small dermal glands, the secretion of which
produces the luminosity when it, by exhaustion from the glands,
comes in contact with the suirounding water. In a few Ostracoda
a brilliant luminosity is produced in a similar way from glands in
the labrum ; it Avas already suggested by G. W. Miiller in 1890,
and has since been observed and published by another author.
Furthermore, I can mention that during a voyage in the Indian
Ocean, Dr. Th. Mortensen met with a vast number of a pelagic
Ostracod which showed a most brilliant light, and he obserAefl
how this was produced. Finally, the present writer has obser\-ed
nejirly the same in a number of a lai-ge Ostracod which had
been procured in Davis Strait. Lute in the evening I observed

76 DR. H. J. HANSEN ON CRUSTACEANS [Jan. 20,

luminoiis points in recently sieved bottom material from
318 fathoms, and undertook instantly some manipulations with
the animals : the luminous fluid came from the head, probably
from the large labrum, and flowed backwards between the shells,
illuminating brilliantly the space between them, especially on the
ventral side. As to the Entomostraca, we can therefore not speak
of " luminous organs " in the common sense of the word, the light
being always produced in the above- described way.

In almost all Euphausiidee real and highly- developed luminous
organs are found, but they differ in structure very much from
those in Serg. challengeri. The reader is referred to Sars's
Report on the ' Challenger ' Schizopoda, and to Chun, " Leucht-
organe und Facettenaugen " (Bibliotheca Zoologica, Heft xix.
Lief. 4, 1896). The highest number of organs met with in this
family is only ten : one on each eye-stalk, two on each side of the
thorax near its inferior margin, and one in the median line of
each of the four anterior abdominal segments. The light is
produced by the " Streifenkorper " (Chun) — "a flabell if orm bunch
of exceedingly delicate fibres, exhibiting in fresh specimens a most
beautiful iridescent lustre" (Sars, op. cit. p. 71) — situated a little
behind the centre of the organ in a mass of large cells ; a biconvex
lens, which is present in the organs with the exception of those on
the eye-stalks, is internal, while the outer chitinous skin is thin ;
a reflector is developed nearly as in Sergestes.

The Danish zoologist Cand. mag. Ad. S. Jensen has directed
my attention to a book by a French author, and lent me a
German translation : ' Die Leuchtenden Tiere und Pflanzen. Yon
Henri Gadeau de Kerville. Aus dem Franzosischen iibersetzt
von W. Marshall, 1893.' In this popular treatise I saw that the
' Talisman ' had captured a deep-sea shrimp with numerous
luminous organs. I attempted in the ' Zoological Record ' and
elsewhere to discover where that animal had been described, but
in vain, and I will therefore reprint the passage in question from
the German book : — " Wahrend der wissenschaftlichen Expedi-
tionen des Talisman fing man in einer Tiefe von 500 m. einen
langschwanzigen Krebs {Acanthephyra pellucida A. Milne -
Edwards), welcher ein lebhaftes Licht um sich zu verbreiten im
stande war und zwar mittels folgender verschieclener Leucht-
organe; erstens befand sich eins am Yorderrand einer Deckschuppe
der Augen, zweitens verlief eine lange leuchtende Linie am
Aussenrand des Tarsus des fiinften Beinpaares, an dessen innern
Basis sowie an der des vorhergehend en Beingiiedes sich weiter
leuchtende Flecke befanden, drittens lag^n ganz ahnliche Flecke
an der Basis des zweiten Gliedes des dritten und vierten Bein-
paares und ebenso je einer an der Basis des Tarsus derselben
Gliedmassen, viertens sah man einen langen Fleck an der Basis am
Endgliede des hintersten Kieferfusspaares, fiinftens verlief ein
schimmerndes Querband iiber die Hiifte des hintersten Thorakal-
f usspaares, sechstens war eine Doppelreihe glanzender Punkte an
jedem Gliede der Aussengeissel der Brustfiisse, sowie an dem.
ausseren Blatt der BaucMiisse vorhanden, siebentens zeigte sich

1903.] IX THE 'CHALLEXGEr' COLLECTION. 77

eiiie (loppelte L'nie leuehtender Puiikte ent.laug der aiisseren
Geissel der luneufiihler nml achtens verlief eine iin hiuteren
Telle zusammeuhaiigende, iiu vordei'eii in Puukte aufgelciste Linie
parallel zum Uutenuude des Riickeupanzei's." The arrangeineut
of tlie organs in Acanthephyra pelliocida is rather different from
that in Sergestes chcdlengerl, but it shows yet more resemblance
to it than to that in the Euphausiidie. Both in *b'. challengeri
and in A. pellucida an astonishingly high number of organs exist,
but as to the structure of the organs in the last-named form
unfortunately nothing is known. I suppose that they are renl
compound organs, not dermal glands aa in the Entomostraca.

If we look for comparison between all luminous animals, it will
1)6 found that only some deep-sea or pelagic fishes and two
Cephalopoda of the genus Jlistloleathis (according to Verany and
Joubin) possess a number of ryal organs which can be compared
with that found in iSe7'g. challengeri and Acanth. pellitclda.

I have looked for luminous oi'gans in all adult species of
Sergestes known to me and in Acaii.ihephyra pLirpn,rea A. M.-Edw.,
but the result was absolutely negative. It is a very curious fact
that about 150 very compound organs are found in one sj^ecies of
Sergestes, while they seem to be quite absent in all other species
hitherto known of the geniis. It may be added that the luminous
species does not deviate from some of the other foi-ms in any other
character of importance : it belongs, even within the genus, to a
group which contains several deep-sea. forms closely allied to e;ich
other. Considered in this light, the existence in one species of
about 150 compound organs seems to me a most astonishing
feature.

Su2)p>lementary Note.

In the preceding section it has been mentioned that among tlie
Cephalopoda two species of HistioteiUlus possess a large number
of compound luminous organs. My frieiuls Prof. G. B. Howes
and Rev. Th. R. R. Stebbing have very kindly dii'ccted my
attention to two papers, which contain some additional knowledge
of the same topic and may be quoted here. W. E. Hoyle (Mem.
& Proc. of the Manchester Liter. & Philos. Soc. vol. xlvi. part vi.
1902) points out and describes the structure of twenty-nine
luminous organs in Pterygioteidhis margaritifera, but this numbei",
though rather high, is yet considei'ably lower than that in
Sergestes challengeri. C. Chun (' Aus den Tiefen des Weltmeeres,'
Jena, 1900) describes and figures (p. 532) the arrangement and
the colours of twenty-foui- luminous organs in '■^ Enoploteuthis
diadema Ch., n. sp." Furthei'more, he writes (pp. 532-33) :
"Ahnliche, wenn audi etwas kleinere Organe, besetzen bei
Vei'ti'etern der Gattung Calllteathis <lie gauze Kiirperobei'llache
von den Armen bis zu den Scliwanziiossen. Die Bauchseite ist
auch liier wieder reichliclier mit ihnen a.usgestattet, als die
Riickenfiiiche." An acconqjanying figui'e of a '' CalUteuthis n. sp.,"
seen from the ventral side, shows a nuinl:er of organs consideraljly
surpassing that in Serg. challengeri.

78 ox CRUSTACEANS IN THE ' CHALLENGER' COLLECTION. [Jan. 20,

EXPLANATION OF THE PLATES.
Plate XL

1. PetalicUum foHaceum Bate. p. 34.

Fig. 1 a. Rostrum of a specimen from Stat. 159, from the side.

1 b. Eye and left antennular peduncle of the same specimen, from above.

1 c. Rostrum of a specimai f/om Stat. 116, from the side; X 5.

1 d. Eyes of the same specimen, from above ; slightlj' more than X 5.

1 e. Basal part of mxp.^ — mxp."* and trl.i — trl.'', together with the branchial
apparatus of the last-named specimen ; X 5. br., rudimentary pleuro-
hranchia to mxp.^; ep., epipod on raxp.2 with its branchia j I., four
pleurobranchial lamella? belonging to mxp.-* and trl.' — trl.^

ly. Lateral view of the front part of the scutum and the eye of a young speci-
men — measuring 21'o mm. in length — from Stat. 158.

Iff. Rostrum and eyes of the same specimen, from above.

2. Fetalidmm sp. p. 55.

Fig. 2 a. Front portion of the scutum of a young specimen from Stat. 137, described
by Bate as Sergestes profundus Bate.

3. Sergestes profundus Bate. p. 69.

Fig. 3 a. Front part of the scutum, eye, and antennular peduncle of the specimen from
Stat. 300; X15/2.
3&. Rostrum of the same specimen.

4. Sevffestes preJiensilis Hate. p. 56.

Fig. 4(7.. Rostrum, eye, and base of the antennular peduncle, from the side.

4 b. Front end of the scutum, ejes, jpeduncles of the antennulie and squama?,

from above.

5. Sergestes Icroyeri Bate. p. 58.

Pig. 5 a. Rostrum, from the side.

5 b. Front end of the scutum, left eye, and the peduncles of the antennulaj, from

above; X 4.

6. Sergestes similis, n. sp. p. 60.

Fig. 6 a. Front end of the scutum, left eye, and peduncle of the antennula, from the
side ; X 14/3.

6 b. Rostrum and supra-ocular spine, from the side; X 12.

6 c. Front end of the scutum, eye, peduncles of the antennulte and squamaj,

from above ; X 5.
Qd. Branchia; above trl.2— trl.-i ; X 11/2. 2br., first branchia to trl.2; SI.,

branchial lamella to tr]."; Sbr., first branchia to trl.'^; 3 br.^, second

branchia to trl.'* ; &c.

Plate XII.

1. Sergestes arcticas Kr. p. 60.

Fig. 1 a. Front end of the scutum, eye, and basal portion of the antennular peduncle,
from the side ; X 9, 2.
1 b. Rostrum and supra-ocular spine of the same specimen ; X 11.

1 c. Brauchiffi above trl.^ — trl.-i ; X 8. The lettering as in fig. 6 d on the pre-

ceding Plate.

2. Sergestes challengeri, n. sp. p. 61.

In the following figures o signifies luminous organs.

Fig. 2 a. Scutum, with the basal parts of the eye-stalk, antennula and antenna ;
X 5.

2 b. Front end of the scutum, basal part of the eye-stalk, and antennular

peduncle; X 17,2.
2 c. Front end of the scutum with the anterior appendages, from above ; X 9.
Of the eye-stalks only the basal ioint remain.s ; of the left squama the
distal half is omitted; on the right squama the two distal ones of the four
luminous organs are seen through the plate.

1903.] ox THE IIAIR-SI.OPE IX MAMMALS. 79

Tig. 2d. Tlii>lias;il ])iirt-< of tlu' tliR'O i)i)stcvi<ir thoracic Icfrs, tlic hvaiicliiipaliovo trl.- —
trl.', ami the lower i)art ot' the two iiiitcrior abiloiiiinal sijiniciits, Iroin the
siilc; X 11. The lettering of the braiichiic a.s in tig. Od on the nveceilin"-
Plate.

2 e. Left niandibli', from helow ; X 13.

2/. Left tirst niaxilliped, from helow ; X 13.

2 (/. Middle part of the tirst maxillijied, from ahove ; X it. o', iipjier provimal
himiiK^iis orjran ; o-, distal luminous organ seen througli the tindt)pod.

2Ji. Left second maxilliped with its ejiijiod and hranchia, from the outer side ;
X 13.

2 i. The three posterior thoracic segments and tlio two anterior ahdominal
segments, from helow, showing 31 luminous organs; X !>. Of the
l)h'opods, only the basal joint — with its lumiiu)us organ — lias heen drawn.

2 k. Sixth ahdominal segment with the hasal ])art of the telson and the uro])()ds,
and the posterior part of the fifth abihmiinal segment with a part of the
pleoi)od ; X 13/2.

2 1. Exterior branch of the left uropod, from the outer side; X 17,2. The
apical part is wanting.

2 m. Luminous organ from the side of the thorax near br. ahove trl.^, seen in
oi)tical vertical section; X 180. a, chitinous lens; a', its thinner
external layer; b, inner lens; c, thin layer between the inner lens and the
thick glandnlar layer d; c, reflector; /", enveloiMiig thin layer.

2«. Sketch of till' luminous organ from the ujiper side of the tirst maxilliped ;
X ISL). ((, lens, jiartly overlapped above bv the iirotiudiiig chitin(*us
plate /.

February 3, 1903.

Howard Saunders, Esq., F.L.S., Vice-President,
in the Chair.

The Secretary read the following report on the additions to the
Society's Menagei'ie during the month of Januaiy 1903 : —

The registered additions to the Society's Menagerie during the
mouth of Januaiy 1903 were 58 in number. Of these 15 were
acquired by presentation and 9 by purchase, and 34 were i'ecei\-ed
on deposit. The total number of dej)artures during the same
period, l)y death and removals, was 120.

The following papers were read : —

1. Notes on the Hair-Slope of four Typical Mammals.
By Walter Kidd, F.Z.S.

[Received November 27, 1902.]

The Otter. Domestic Dog, Ox, and Horse have been selected
for consideration as showing very different arrangements of tlieir
hairy coverings, and as affording by their environments and habits
the probable explanation of the differences found. Two CarniNores
and two Ungulates are thus compared and contrasteil.

(1) In the Otter, taken as a type of the long-liodied hairy
mannnal with very short limbs, the liair presents an unbi'okeii
slope from the snout to the tip of tlic tiil. On the liead and

80 DR. W. KIBD ON THE [Feb. 3,

trunk it passes from cephalic to caudal, and on the limbs from
proximal to distal exti'emities in a pei-fectly tiniform manner.
This ai-rangement obtains in a great number of long-bodied or
primitive forms, such as rodents, smaller carnivores, insectivores,
marsupials, lemurs and lower monkeys. As to the hair-slope in
the Otter, nothing further requires notice, and the type may be
looked upon as that of the primitive hairy mammal.

(2) When a Carnivore of a difFei-ent form, such as a short-bodied
close-haired Domestic Dog, is investigated, a few departures fiom
the primitive type of hair-slope are found. The Dog is taken as
representing a more highly developed carnivore form, and it
presents, as do all Canidfe possessing sufficiently short hair : — ■

(i) A whorl situated at the extremity of the snout with a
feathered arrangement proceeding from it towards
the orbits,
(ii) Reversed bilateral area of hair on the pectoral region,
(iii) Reversed bilateral area of hair on the ventral surface of

the abdomen in the inguinal region.
(iv) Bilateral symmetrical whorl in the gluteal region lying
exactly over the tubera ischii, and, proceeding towards
this spot, a reversed area of hair on the extensor
aspect of the thigh.
(v) Revei'sed area of hair on the extensor surface of the
ulna.
These have been fully described, and explanations of their
setiology put forward, in our Proceedings \

The Domestic Dog thus presents an advance in several directions
upon the primitive arrangement of hair in the Otter.

(3) The Domestic Ox shows certain peculiarities and departures,
both from the primitive and specialized Carnivore type : —

(i) In the frontal region a whorl and a feathering which
proceeds from it towards the level of the external ears,
terminating hei'e in a crest.
(ii) Short longitudinal crest or mane, commencing as a tuft

at the level of the horns,
(iii) This crest or mane is interrupted abruptly in its normal
backward course at about the middle of the neck by a
ridge. At this point there is produced a meeting of
two streams of hair pi'oceeding in opposite directions,
and the forward or " abnormal " stream is found to
commence at a whorl which lies in the median plane
at about the middle of the dorsal region.
(iv) From the whoi'l situated on the dorsal region the slope
of hair resumes its normal direction and passes to the
tail, on Avhich it is so ai'ranged that a central longi-
tudinal crest is usually found, and it terminates in a
scanty tuft of hair.

1 Proc. Zool Soc. 1902, vol. ii. pp. 148, 149, 150.

1903.] HAIR-SLOPE IX MAMMALS. 81

The Ox presents no pectoi-al whorl as do the Dofj and Ilcn-se, tlie
dewlap in this region being covered with perfectly smooth and
uniformly arranged hair.

These are the only four constant peculiarities of hair-slope, but
occasionally the flank and post-humeral regions, and extensor
siTrface of the thigh, show small whorls and featherings connected
apparently with the action of the panniculus carnosus muscle.

(4) The Horse is more specialized in the matter of hair-slope
than any other animal except man.
It shows constantly — •

(i) Frontal whorl, feathering, and crest.

(ii) Tuft between the external ears,
(iii) Longitudinal crest or niiine.

(iv) Strong efiicient tail with bushy hair, reaching to the
fetlock.

(v) Inguinal whoil, feathering, and crest,
(vi) Pectoral whorl, feathei'ing, and crest.

Frequently but not constantly : —

(vii) Post-humeral whorl, feathering, and crest,
(viii) Cervical whoi'ls, featherings, and crests in several
situations,
(ix) Gluteal whorl, feathering, and crest,
(x) Tuft on lateral aspect of the abdomen.

In passing from the simple type found on the Otter to that of
the Horse, a very significant series of changes is thus displayed.

Such facts as are here referi-ed to have little or no intrinsic
interest or importance, but the phenomena of Nature, small and
great, demand explanation in accordance with the methods of
science, and it is impossible to ignore the peculiarities of hair-
slope which have been taken here as typical of a vastly greater
number in other animal forms. Any other interest they may
have is subordinate to their relation to the problems of heredity.
The four groups of facts suggest apparently two explanations as
to their (etiology. The first and most obvious is that some of them
ai'e adaptive modifications of value to the animal ; the second,
that others are produced by its habits ; and it is not difficult to
distinguish these two classes in the four typical forms chosen for
consideration,

1. In the Otter the uniform trend of hair requires no other
explanation than that this arrangement of hair ofters the least
possible resistance to mo^■ements in the water and in burrows.

2. In the case of the Domestic Dog the departures from a primi-
tive type can hardly be ascribed to anything else than to use or
habit : they are adapted hy the habits of the animal, not for its
needs.

Proc. Zool. Sor._1903, Vol. I. No. VI. 6

82 DR. W. KIDD ON THE [Feb. 3^

3. The Domestic Ox presents cei-tain points of interest intimately
associated with adaptive modifications, but many of the phenomena
are evidently not so associated. Thus, along the median line of
the dorsal aspect of the Ox is seen a tuft, crest or mane, whorl,
and long efficient tail, the length of the last being such as to
reach beyond the dorsal whorl. The large hairy external ears,
which can be flapped backwards so as nearly to reach the middle
line, may also be included. These may be looked upon as adaptive
modifications, existing for the purpose of defending the animal
against injuiious insects; and a few observations on several oxen
and horses have been made as to the importance and fi'equency of
these, and as to the use of the panniculus carnosus, even in this
temperate climate.

In moderate summer weather and an exposed wind-swept
situation, the number of occasions on which certain oxen and
horses flapped backwards and forwards their ears, corriTgated their
skins by the action of the panniculus, and flicked their tails
on to their backs was observed and noted with the following
results : —

Tails flicked by oxen at the rate of 348, 468, 504, 540, 720,

780, 1082 times in an hour.
Ears flapped by oxen at the rate of 684, 816, 840 times in an

hour.
Panniculus acted in oxen at the rate of 984 times in an hour.
Tail flicked by horse at the rate of 1108 times in an hour.

A rough idea may thus be gained as to the importance of certain
of these mechanisms for defence against injurious insects in hot
countries, if in a temperate climate and exposed situation so
frequent a use is made of them.

4. The Horse also presents in the median plane several similar
modifications, as a tuft, mane, and long bushy tail, which, when not
docked, reaches almost exactly to the spot where the mane termi-
nates. The panniculus in the Horse is much moi-e active in that
part of the flank and in the forequai'ters where the tail does not
i-each than elsewhere.

It remains only to point out the distinction maintained here
between modifications of hair -arrangement which are themselves
adaptive, as in the Otter, and others which are, so to speak, by-
products of habits of the animals exhibiting them, such as whorls,
featherings, and crests. The only meaning properly assigned to
adaptive modifications is that such modifications are adapted for
the needs or comfoi't of an animal. When they ai'e adapted hy
the habits of an animal, and have no thinkable relation to its
needs, they must be classed stiictly as non-adaptive phenomena.

It should be further stated that all the phenomena i-eferred to
are congenital, and would seem to have an intimate connection
with the problems of heredity.

1903.] nAIR-SLOPE IX MAMMALS. 83

List I.
Otter No modifications of primitive type of liair-slope.

Dog 2 whorls itc.

3 reversed areas of hair.

Ox 2 whoilsctc.

Tuft and small mane.
Crest and tuft on tail.

Horse ... 3 whorls i-c. ^

Tuft. \ ,, , .

nr - Constant.

Mane.

Bushy tail. J

4 whorls etc. Occasional.

List II.

OXTKR.

Dog....
Ox ....

HoKSB.

T, J Phenomena of

JJepartiires Tr •

J" It ■ -L- Hair-arranue-

froin Prinntive I , 7 .■'

■' ,r . , tinent, adaptive or

•^ non-adaptive.

None.

All adaptive.
All non-adaptive.

J^nvironments.

I faint. I.

4 constant, several I non-adaptive,
occasional. 1 3 adaptive.

Non-adaptive.

I
fi con.stant, 4 oc-! 3 non-adaptive,
casional. j 3 adaptive.

Non-adaptive.

Homogeneous. , Simple.

More varied. More varied.

I
Still more varied. Still more varied

Most varied of
the ibur t^pes.

Complex, com-
binini; tlieliahiti
of its wild lift
witli those of
domestication.

List III.

Observations as to use of E.\:ternal Ears, Panniculus
Carnosu.s, and Tail in Ox and Horse.

OxEX flapped external ears at the rate of 684, 810, 840 times

an hour.

In OxEX panniculus acted at the i-ate of 984 times an houi\

OxEX flicked tails at the rate of 348, 468, 504, 540,

720, 780, 1082
times an hour.

Horse flicked tail at the rate of 11 08 times an hour.

84 CATT. F. WALL OK THE [Feb. 3,

2. A Prodromns o£ the Snakes hitherto recorded from China^
Japan, and the Loo Choo Islands ; -with some Notes.
Bj Captain F. Wall, Indian Medical Service.^

[Received December 15, 1902.]

"Whilst attached to the China Expeditionary Forces from
1900-1902, I had opportunities of examining the Snakes pre-
served in the three Museums in China, and also others in private
collections. My notes on these, together with those on specimens
obtained myself, form the subject of this paper, which I have
arranged so as to form a complete prodromus of the species
hitherto recorded from the countries above specified.

In spite of the large aggregate of specimens I examined, it is
perhaps worthy of remark that I failed to discover one species
new to science'", and this only serves to show how extensively and
thoroughly this branch of natural history has been worked out.

In the City Hall Museum in Hongkong, out of about one
hundred specimens from the territory above mentioned, I found
many misnamed and others unidentified. I was informed that
during a typhoon some years previously a large case, containing
specimens, was blown over and, the contents wrecked. Out of
the debris labels were recovered as far as possible and replaced,
but some were evidently incorrectly reattached and others were
destroyed. This circumstance may render the accuracy of some
of the records open to question. That this collection is far from
representative is evidenced by the fact that during five and a half
months' residence in this Port I obtained six species which were
not to be found in the Hongkong Museum.

In the Museum in Shanghai I found about fifty specimens
which were for the most part old, and concerning which there
was practically no information regarding their habitat.

In the Museum belonging to the Jesuit Fathers at Siccawei,
near Shanghai, I examined some two or three hundred specimens,
but these, again, furnished practical!}^ no record of their habitat ;
however, the late Pere Heude informed me that they had all been
collected in the Yangtse Yalley, and he mentioned Ning-ko-foo
(which I find is on the southern bank of the Yangtse River) as
being the most northern limit. He could not define limits either
to the south or west.

As regards habitat, where I do not cite the authority, so far as
the territory this makes reference to is concerned, it is to be under-
stood that examples exist in the British Museum Collection at
South Kensington.

I have only made reference to points in these specimens which
do not absolutely agree with the desci'iptions to be found in

1 Communicated by G. A. Boulengee, F.R.S., V.P.Z.S.

2 The species referred to hereafter as Dipsadomorphits kraepeUm, No. 65, which
I had hoped to have the honour of describing for tlie first time, was recorded by
Stejneger shortly after I began to prepare this paper .

1903.] SNAKES OF CniXA AXD JAPAN', 85

Mr. Boulenger's work ' Catalogue of .Snakes in the British
Museum,' 1893-1896, and 1 have also adopted the nomenclature
of that woi-k.

I am indeVjted to Mr. Boulenger for much valuable infoi-mation
and assistance in drawing up tliis synopsis. My tlianks are also
due to Mr. F. Bowley, Hongkong, Dr. Barchet, Shangliai, and
the late Pere Heude, 8iccawei, for their courtesy in granting me
access to the collections in the Museums under their super-
intendence, as well as to Mr. A. Owston, Yokohama, and
Mr. Armsti'ong, Hongkong, for placing their specimens at my
disposal.

Part I. — List of Chinese OniiDiA,

TyPHLOPIDyE.

1. TvPHLOPS LiNEATUS. Maliiy Peninsula and Archipelago,
Hongkong ?

2. TvPHLors BRAJiiNUS. Southei'u Asia from Ai'ahia to Malay
Archipelago, South China, Formosa, Hongkong, Hainan, Mau-
ritius, Comoro Islands, Madagascai-, Cape of Good Hope. — I
obtained one small specimen three and a half inches long on the
Peak in Hongkong Island.

BOID/E.

3. Python reticulatus. Burma, Indo-China,, Siam, Malay
Peninsula and Archipelago, S. China, Hongkong {City Hall
Mus.). — I saw three specimens, one in the City Hall Museum,
Hongkong, labelled from that island, one in the Shanghai
Museum, habitat unrecorded ; and one in the Siccawei Museum,
consisting of a head only. From the information given by the
late Pere Heude concerning this collection, I must include this
species among the Chinese fauna. In all the specimens there are
three rows of priefrontals, and the lower labials from the 2nd to
4th and from 13th to 18th are pitted. The frontal is divided
longitudinally in two of the specimens. A^entrals 315-319.
Subcaudals 92-100.

4. Python molurus. Southern Asia from India to China,
Hainan, Hongkong {City Hall Mas.), Formosa {Sivinhue).—
The Chinese name given me was " Hiang zo," or " aromatic
snake," perhaps in reference to their cooked flesh, which is highly
esteemed by these people. In Formosa it is called " Uang '
i^Swinhoe).

COLUBRID.E.

5. AcROCHORDus JAVANicus. Malay Peninsula, Java, New
Guinea, Hongkong {City Hall Mvs. No. 293).

6. PoLYODONTOPnis coLLARis. Himalayas as far west as Simla,
Assam, Arrakan, Upper Burma, S. China, Hongkong. — I ob-
tained one specimen on Hongkong Island. There is one anterior

^6 CAPT. K. WALL ON THE [Feb. 3,

temporal on both sides, wbicli comes in contact with the eighth
labial only ; this latter is unusually high and is probably the
result of a confluence with the normal inferior anterior temporal.
The first lower labials do not meet behind the symphysis. Belly
unspotted,

7. Tropidonotus savinhonis. Formosa.

8. Tropidonotus nuchalis. China.

9. Tropidonotus vibakari. Manchuria, Japan, Formosa.

10. Tropidonotus piscator. Southern Asia from India to
China, Hainan {Herz & City Hall Mus.), Hongkong {Hallow. &
City Hall Mus.), Formosa \Stejneger). — Apparently common in
the extreme south. I obtained three specimens from the main-
land opposite Hongkong, and saw one in Mr. Armsti'ong's col-
lection which he assured me he procured from Hongkong Island.
I found only two specimens in the large Siccawei Collection. In
one specimen (froan Kowloon, opposite Hongkong) there ai-e five
postoculars on one side, in another four on both sides. In one
there are eight upper labials, with the fourth only touching the
eye on both sides.

11. Tropidoxotus annularis. China, Formosa. — This must
be a very common snake in the Yangtse A^alley, judging from the
large numbei- of specimens in the Siccawei Museum, but in the
extreme south it appears to be rare or absent. There are two
prfeoculars on both sides in one specimen, five postoculars on
both sides in one specimen, and two postoculars on one side
in one specimen. Ventrals 132-164. Subcaudals 70 in one
specimen.

12. Tropidonotus tigrinus. Siam, Cochin China, China,
Hainan, Hongkong ? Manchuria, Corea, Japan. — Called " Yeh-
chi-po" by the Chin :se, signifying pheasant's neck, also " Ch'ing-
ch'ang-chung {Moll.). Apparently as common as T. annularis
in the Yangtse Valley, from the number of specimens at Siccawei.
One specimen in Mr. Armstrong's collection was, he informed me,
captured on Hongkong Island ; however, it was not labelled, and
his collection contained some species he had procured in Japan.

13. Tropidonotus stolatus. Southern Asia from India to
China, Hainan, Hongkong, Formosa, Chusan Archipelago, Philip-
pines. — I obtained one specimen from the mainland opposite
Hongkong. The labials on one side were nine in number, and
the fourth, fifth, and sixth touched the eye. I saw no specimen
in the Shanghai or Siccaw^ei Collections.

14. Tropidonotus subminiatus. Eastern Himalayas, Assam,
Burma, Malay Peniasula and Archipelago, S. China, Hongkong.

15. Tropidonotus chrysargus. Eastern Himalayas, Assam,
Burma, Malay Peninsula and Archipelago, S. China, Hainan.

1903.] SXAKCS OF CIII.VA AXD JAl'AX. 87

16. Tropidoxotus balteatus (Cope), Proc. Ac. Phil;ul. 1894,
p. 426. H:tiiian.

17. Tropidoxotits craspedogaster (Blyr.), 1^. Z. S. Loml. 181)9,
p. 163, pi. xvii. fig. 1. China.

18. Tropiuoxotus percarixatus (Blgr.), P. Z. 8. Load. 1899,
J). 163, J)!, xvii. fig. 2. China. — I saw six specimens in tlie
Siccawei Collection which exactly fit the description of Mr. Bou •
lenger's specimen except in the following details: — Length of
internasals sometimes eipials pi-jefrontals. Postocnlars and sub-
oculars : the numbers of shields in contact with the back of tlie
eye, intervening between the supraocular above and the fifth
labial, are four in two specimens on both sides, four in one
specimen on one side, five in two specimens on both sides, fi\e
in two specimens on one side, six in one specimen on one side.
Labials eight with third and fourth touching the eye on one
side, nine with fifth only touching the eye on one side. Ventrals
and subcaudiils 140 + 73, 139+ ?, 139-^73, 138+ ?, 138 + 72 (tip
slightly docked).

19. Tapixophis LATOucnn (Blgr.), P.Z. S. Lond. 1899, p. 164,
pi. xviii. figs. 1-1 c. China.

20. PsEUDoxEXODOX MACROPS. Him:d'\yas, Khasi Hills, Hills
in Biu'ma, Yunnan [de Scahra^), and S.W. China.

21. PsEUDOXEXODOX DORS ALLS. China.

22. Opisthotropis axdersoxii. Hongkong, — I obtained five
specimens, all from Hongkong Island. They were all captured in
a swamp near the Sanatorium on the Peak, whilst being drained
during the campaign against malai-ial mosquitoes. One was dug
up at a depth of about 2 feet below the surface. They accord
with Mr. Boulenger's description except in the following parti-
culars : — Labials are inconsistent in ai-rangement. There are nine
in one specimen on one side, eight in all the rest. The fourth
and fifth touch the eye in one specimen on one si<lc, the fifth
only touches the eye in two cases on both sides, and in one case
on one side, no labials touch the eye in two specimens on one
side. Pmeoculars (including the suboculars of Boulenger) are
two in two specimens on both sides, two in two specimens on one
side, one in one specimen on one side, and three in one .specimen
on one side. The anterior chin-shields are in contact with five
lower labials in two specimens on both sides, and in one specimen
on one side, with four in one specimen on both sides, and in
one specimen on one side. Ventrals and subcaudals 167 + 60,
149 + 53, 164 + 59, 165 + 59. Colour uniform dull olive-bluish
above, lower half of ultimate row and belly yellow. Lower labials
and throat-scales with dull bluish mottling. Sparse mottling
beneath tail.

1 JJull. Mus. ir. N. Paris, 1«>7, iii. V- 215.

88 CAPT. F. WALL ON THE [Feb. 3,

23. Trirhinopholis styani (Blgi\), P. Z. S. Lond. 1899, p. 164,
pi. xviii. figs. 2 & 2 a. China.

24. AcHALii^us RUPESCENS. Hoiigkong. — I obtained four
specimens in Hongkong Island, found in the low vegetation on
the slopes of the Peak. In one specimen both anterior temporals
touch the eye on both sides. Scales somewhat irregular, 23-25
in mid-body. Yentrals and subcaudals 150 + 56, 158 + 58, 158 +
61, 154 + 58. Colour uniform olive-brown above, slightly darker
vertebrally, and with iridescence in reflected light. Head same
colour above, merging to chestnut on temporal regions.

25. AcHALiNus BRACCONiERi. S. China. — I think that this
species will have to be united with A . spinalis ; the differences
between the two shown by Mr. Boulenger ^ are considerable but
not constant, being shared by individuals of both supposed species.
For instance, in all the four specimens of A. spinalis that I have
examined (one of which I presented to the British Museum,
which has been seen and identified as such by Mr. Boulenger)
the scales are 23, and not 21. The relative length of the sutures
between the internasal and prsefrontal has, in my opinion, little
or no weight. The specimen of A. spinalis 1 sent to the British
Museum has the internasal suture about two-thirds the length of
the praefrontal ; and in more than one specimen of hracconieri in
the British Museum a similar condition exists. The specimen
figured by Stejneger ^ from Japan which he calls spinalis is more
like Mr. Boulenger's hracconieri, but has 23 scales.

26. ACHALINUS SPINALIS. China, Japan.- — I saw three speci-
mens in Mr. Owston's collection from Mount Fuji, Japan, and
one I found in the Siccawei Collection. In all there is a large
shield on the postero-lateral region of the parietals similar to
those described under ^4. h-acconieri, and, like them, separated by
one scale in the median line. Ventrals and subcaudals 154 + 58,
165 + 44, 165 + 48, 170 + 44.

27. Lycodon aulicus. Southern Asia from India to Malay
Archipelago, Philippines, Formosa {City Hall Mus.), Hongkong?
{Boettger ^), S. China ? viz. from Amoy {Steindachner *).

38. Lycodon fasciat^us. W. Yunnan [Anderson), Assam,
Burma.

29. Lycodon subcinctus. Malay Peninsula and Archipelago,
Sumatra, Borneo, Java, Timor, Hongkong, Philippines. — There
are two specimens in the City Hall Museum from Hongkong and
one from Timor. I also examined a fourth, which Mr. Armstrong

1 Cat. Snakes Brit. Mus. vol. i. pp. 308 & 309.

2 Ann. Zool. .Tapon. ii. p. 29.

3 The specimen alluded to by Boettger (Mat. herp. Faun, von China, 1888, p. 84)
in the City Hall .Museiiin is no longer in the collection; the only specimen of this
species in that In^stitution is i'rom Formosa.

'' Eeise der Novai-a, Rept., Wien, 1869, p. 74.

1903.] SXAKES OF CUIXA AND JArAX. 89

informed me he had captured in the filter-beds near Bowen Road
on Hongkong Island. In two specimens the third, fourth, fifth,
and sixth supralabials touch the eye on both sides. Ventrals
192-209 \

30. DixoDOX RUFozoNATUS. China, Hainan, Formosa, Chusan,
Corea, Tsu Shima, Japan, Loo Choo Islands. — This is a very
common snake about Shanghai, whei-e I encountered it frequently.
I picked up several and allowed them to ci-awl about my arms
without their exhibiting the least malice or alarm, and they
made little attempt to escape in the first instance. Riding my
bicycle one night I saw one crossing the road ; I dismounted,
flashed my lamp on the snake, and, while holding my machine
Avith my light hand, captured it easily with my left. It made no
attempt at escape, though cover was within a yard or two. It
Avas full-grown and not des(]unmating. My servant caught one
in camp one night just outside my hut ; it encircled the man's
leg but did not bite. It contained a large toad {Bufo vulgaris).
I found a large number preserved in the Siccawei Museum. I
found one in Mr. Owston's collection obtained from Japnn, and
four othei-s procui'ed fi'om Ishigaki Island in the Loo Choo group.
I noted the following : — Internasals half or less than half the
length of pr^efrontals. Loreal, in Japanese and Loo Choo speci-
mens, does not touch the eye in all (five) ; in Chinese does not
touch the eye in five, touches eye in eleven. Postoculars three
on both sides in one specimen, three on one side in one specimen.
Temporals one on one side in one specimen. Labials normal in
all. Anterior chin-shields in contact with four lower labials in
one specimen on both sides, with six lower labials on one side
in two specimens. Ventrals in Japanese and Loo Choo speci-
mens 180-190; in Chinese specimens 192-209. Subcaudals iia
Japanese and Loo Choo specimens 76-87 ; in Chinese specimens
64-76. Colour : there appear to be two very distinct varieties.
All the Chinese conform to the following description : — Alternate
bars of jet-ljlaek and coiul-i'ed (white in old spu-it-specimens)
doi'Siilly, breaking into a coai'se mottling on the flanks. The black
bars involve two or three scales in the length of the snake, and the
red one scale or slightly moi-e. Thei'e ai-e 53-74 black bars on
the body and 18-24 on the tail, the first is broadest and forms a
chevron on the nape. Head black, fading to wdiitish on labials ;
sutures on crown coral-red. A light temporal streak usually.
Belly whitish, with some lateral mottling. The Japanese and
Loo Choo specimens agree : — Alternate darkish-bi-own and dirty
whitish (perhaps red when fresh) bars doi'sally, breaking up into
a mottling laterally. The bi'own bars involve four or five scales
(more quite anteriorly), the light one scale. Thei-e are 24-33
brown bai's on body, 15-18 on tail. Crown of head broAvn, fading

1 Giinther records also Li/codnn (OpJtilcs) alLofiiscus from Formosa (Ann. Mafr.
Nat. Hist. (4) vol. i. 180S, p. 420), but jjives no dcscni'tion nor authority for
recording it.

90 CAPT. F. WALL ON THE [Feb. 3,

to whitish on labials. A light temporal streak. Belly whitish,
with or without sparse lateiul mottling, except beneath tail
where this is abundant.

31. DiNODON SEPTENTRiONALis. China, Formosa, Himalayas?

32. Zaocys DHUMNADES. S. China, Chusan. — A very common
snake about Shanghai and evidently throughout the Yangtse
Valley, for there are many specimens in the Siccawei Museum.
I have encountered it frequently, and consider its generic name
most apt, as it is veiy active, swift, and clever to elude capture.
I saw as many as four in a day's ramble in the spring, when
batrachians wei-e clamouiing in the water engaged on matrimonial
matters. The snakes wei-e each coiled vip on the banks close to
the water, and in low vegetation, awaiting the excursion of some
unwary individual. One I saw coiled up similarly two yards or
so away from two toads {Biifo vulgaris), Avhose curious behaviour
it was that first attracted my attention. In spite of every cai-e
my presence was always detected by the snake before I was
aware of its presence, and it immediately made off in great haste
for the water, and disappeared among the i-oots of the aquatic
vegetation. On one occasion in the summer I watched one for
some time in a strip of grass in the open, myself unobserved ;
and it was most interesting to notice the method and care
with which it beat the patch of gi-ass like a hai-rier, prying into
every recess in the ground or tussock that might harbour some
possible prey. I have seen Chinese jugglers wdth this snake in
their stock in trade, and I believe that it is this species that
is even now occasionally met with in houses in the heart of
the town of Shanghai. On October 3rd, 1901, I captured a
young one recently hatched in camp which closely resembled the
adult in colouring. Loreal : in one specimen there are two
superposed shields on one side. Temporals : a single anterior
in one specimen on one side. Anterior chin-shields in contact
with four lower labials on both sides in one specimen. Ventrals
190-199. Subcaudals 96-119.

33. Zamenis korp^os. Sikkim Himalayas, Assam, Burma,
Siam, Malay Peninsula, Sumatra, Java, W. Yunnan, S. China,
Hainan, Hongkong, Formosa {City Hall Mus.). — I obtained one
specimen on the mainland opposite Hongkong. There are two
specimens in the Siccawei Museum. Loreal single on both sides
in one specimen, three on one side in one specimen. Subcaudals
100-116.

34. Zamenis mucosus. Southern Asia from Transcaspia to
Malay Peninsula, Java, S. China, Hainan {City Hall Mas. <& Herz),
Hongkong, Formosa, Chusan {Cantor d' Gilnther). — I obtained
two examples from Stonecutter's Island in Hongkong Harbour,
and three from the mainland opposite Hongkong. There are two
specimens in the Siccawei Museum. Labials nine, the fifth and
sixth touching the eye in one specimen on one side.

1903.] SXAFvES OF C'lilXA AXD JAPAN'. 91

35. Zamexis snxALis. Siani, H;iin:in, China, Coroa, IMonijolia.
This snake is e\'idently veiy common in the Yangt.se Valley, as
there are ten specimens in the Siccawei IVIuseum. Concerning
one which I obtained from Huangtsun, N. China, my donor wrote
me that he had found it in his verandah, it having dropped, lie
believed, from the i-oof abovit 7 feet above. He discovei'e(l it in the
act of devouiing a lizai'd [Crecko suhpalmatas). It had seized the
gecko by the body and was encircling it in its coils. A little later,
Avhen the captive had exhausted its futile struggles, the snake
seized it by the snout and began to swallow it. Frontal three-
fourths to foui'-fiftlis parietals. Loreals : two (anteiior and
posterior) on both sides in one specimen. Subocular absent in one
specimen on one side. Praeoculars : the upper touches the frontal
in two specimens on both sides. Tempoiuls : two anterior in all
specimens. Labials : eight, with the fourth and fiftli touching
the eye in two specimens only, the normal ariungement being
nine, with the fifth and sixth touching the eye. Antei'ior chin-
shields in contact with four lower lahials in two specimens on both
sides. Ventrals 179-207. Subcaudals 84-100.

36. Coluber porphyraceus. E. Himalayas, Hills in Assam,
Burma, Malay Peninsula, Siuuatra, Yunnan {Anderson).

37. Coluber maxdarixus. Chusan, China (N.W. Fokien,
Blgr. ; Prov. Chihli, Mijll.). I saw one specimen in the Siccawei
Collection. It is called by the Chinese " 'Hua-tai-tsze," meaning
variegated girdle {Mull.).

38. Coluber rufodorsatus. E. Siberia, China, Chusan, For-
mosa, Hainan. — Evidently a very common snake. There are fi\e
in the Shanghai Museum, and a large number at Siccawei.
Frontal often greater than distance to end of snout. Anterior
chin-shields in contact with four infralabials in one specimen out
of fifteen examined.

39. Coluber diox'E. S. Russia, Transcaucasia, Temperate Asia,
China, Hainan, Japan [Blgr.). — Evidently a veiy counnon snake
in the Yangtse Valley, as the Siccawei Collection contains a large
number of specimens. I obtained two from an officer in Ching-
wang-tao, N. China, who told me the snake was common there.
One specimen contained the brood of some small bird, four in
number, one of which was still partly env-eloped in its shell.
Frontal often greater than its distance to end of snout, sometimes
equalling parietals. Pra-oculars : three on one side in one speci-
men. Postoculars : one on both sides in one specimen out of
sixteen examined. Labials nine, with fourth and fifth touching
the eye on one side in one specimen ; nine with fifth and sixth
touching the eye on one side in two sjiecimens only. Anteiior
chin-shield often greater than posterioi', in contact Avith six
'infralabials on both sides in one specimen, and on one side in one
s})ecimeu. Scales in mid-body 23 in three specimens, 25 in ten,

92 CAPT. F. WALL ON THE ' [Feb. 3,

and 27 in one specimen. This species is called " Huang-ch'ang-
ch'ing " by the Chinese [IfdlL).

40. Coluber t^niurus. Sikkim, Cochin China, Siam, Malay
Peninsula, Borneo, Sumatra, China, Formosa {/Senckenberg Mus.,
Boettg.), Chusan, Manchui-ia. — This snake is common in the
Yangtse Valley. I encountered one near Shanghai, which, however,
I failed to captui'e as it took to water and disappeared in the
submerged vegetation, Loreals confluent with prfefrontals on both
sides in one specimen. Temporals : single on both sides in one
specimen, and on one side in one specimen. Labials : nine with
the foui-th and fifth touching the eye in one specimen on both sides.
Ventrals 226-242.

41. Coluber schrenckii. Amoorland, Corea, N. Japan.

42. Coluber phyllophis. China. — I saw seven specimens in
the Siccawei Museum, the largest measuring 8 feet 1 inch. There
is also a stuffed specimen in the Shanghai Museum. Scales 21
in mid-body in one specimen. Ventrals 221 in one specimen.

43. Coluber davidi. China {Scmvage).

44. Coluber mcellendorffii. China {Ilerz d; Broeclcelmann).

45. Coluber melanurus. Burma, Malay Peninsula, Borneo,
Sumatra, Java, S. China.— I saw one specimen in the City Hall
Museum from Java, Subcaudals 88.

46. Coluber radiatus. E. Himalayas, Bengal, Assam, Burma,
Cochin China, Malay Peninsula, Sumatra, Java, S. China {Miill.,
fferz, Moll., Kaufm., Broeckl.), Hongkong {City Hall Mus.). —
There are five specimens in the City Hall Museum from Hong-
kong. I also obtained one specimen from that island which
contained four blind and callow offspring, the brood of some small
mammal, pi'obably a rat \

47. Dendrophis pictus. S. Asia from India to Malay Archi-
pelago, Cliina {Giinth.), Hongkong, Philippines. — One specimen
of this snake in spirit was given me by a lady, who assured me
she had obtained it on Hongkong Island.

48. SiMOTES purpurasoens. Siam, Cochin China, Malay
Peninsula, Sumatiu, Borneo, Java, S. China ~.

49. SiMOTES CYCLURUS. Bengal, Assam, Burma, Siam, Cochin
China, Malay Peninsula, Sumatra, S. China.

50. SiMOTES FORMOSANUS. S. China, Formosa,

51. SiMOTES viOLACEUS. Bengal, Assam, Burma, Camboja,

-^ Coluber climacoplwrus vel SlapJiis virgatus, described by Boettger (Mat. hei'p.
Faun, von China, 1888, p. 72) from Corea, is, 1 think, an error. The specimSn
alluded to is evidently Coluber schrenckii.

2 pam^ril & Bibrpii, 'Erp^fcologie,' p. 632.

1903.] sxAKES orcnixA AND jArAx. 93

B. Cliina, Hainan, Hongkong. — I saw one specimen in tlie City
Hall Musenni, and obtained two myself on the mainland opposite
Hongkong. One was caught in grass on the hillside and suffeved
itself to be handled without I'etaliiition. The other was caught on
a shi'ub poised over a flower. Frontal : length greater than dis-
tance to end of snout in three specimens. Labials : six with the
third and fourth touching the eye on one side, seven with the
fourth and fifth toucliing the eye on the other side in one speci-
men. Nasals semidi\ided on both sides in one specimen.

52. SiMOTES CHixEXSis. China, Hainan. — Rostral : visible
portion seen from above equals distance to frontal in two
specimens. Labials seven, with the fourth only touching the eye
on both sides in one sj)ecimen. Subcaudals 50 in one and 51 in
another specimen. I saw two specimens in the Sicca wei Museum,
and obtained one myself fi'om Kiangyin on the southern bank of
the Yangtse River, to the noi-th-west of Shanghai.

53. SiMOTES vAiLLANTi. China (Sauvage).

54. Ablabes major. China, Hongkong [Gunth., IlaUow., City
JIall Mits.), Formosa, Chusan Archipelago. — Evidently fairly
common in the Yangtse Valley. I saw one specimen of this snake,
belonging to ]\Tr. Styan, in Shanghai, and many others in the
Siccawei Collection. Thei'e are thi'ee in the City Hall Museum,
two of which ai'e fi'om Formosa, the other of nnceitain habitat.
I saw one specimen in Mr. Armstrong's collection in Hongkong
obtained from that isLuid, whei'e, as he told me, it is not
uncommon. Labials : in one specimen seven, with the third
and fourth touching the eye on one side, Subcaudals 89 in one
specimen, and 90 in two others.

55. Ablabes dori/E. Kachin Hills, Assam, China. — I saw one
small specimen 8 or 9 inches long in the Siccawei Collection.
Yentrals 160 ? Subcaudals 85 ? It agrees in every particular
with Mr. Boulengei-'s desci'iption. This species has not been
previously recorded from China,

56. Calamaria pavimextata. Burma, Siam, Cochin CliiiL-i,
Malay Peninsula, Java, S, China {Moll.), Riu Kiu Archipelago
[Stejneger ').

57. Calamaria berezowski, S. China {Gilnth. ^).

58. Calamaria septextrioxalls. China, Chusan Archipelago,
Hongkong. — I saw two specimens in the Sicwxwei Collection, both
quite typical.

59. SPAXiornoLLS souliei. Yunnan {de Scabra ^).

^ Stejneger (I'roc. Biol. Soc. Wasli. xiv. p. 191) ilcserihes a snake .is new to scieiioe
under the name of Calamaria pfejfcri, which, in my opinion, is u specimen of
C pitvimentdla.

■ Ann Ae. St. Petersb. 1896. p. 205, pi. i. fig-, a.

^ Bvill. .Mas. N. II. V:\v\%, iii. 18S)7, p. 211!.

94 CAPT. F. WALL ON THE [Feb. 3,

60. HypsiRHiNA PLUMBEA. Burma, Inrlo-China, Malay Penin-
sula and Archipelago, S. China, Hainan, Hongkong, Formosa.— I
saw two specimens in the City Hall Museum, one labelled China
and one Hongkong, and I obtained four myself on the mainland
opposite Hongkong. All my specimens were caught in or about
a sluggish stream near our camp at Kowloon. Frontal : length
greater than distance to end of snout in six specimens. Anterior
chin-shields in contact with six infralabials in two specimens on
one side. I noticed that in some specimens the lateral scales
about the anal region had minute central tubercles, reminding
one of the condition in Aspidura cojyii and A. trachyprocta.

61. Hyfsirhixa enhydris. India, Ceylon, Burma, Siam,
Cochin China, Malay Peninsula, Borneo, Celebes, S. China,
Hongkong.

62. Hypsiriiina bennetti. China, Hainan (Ilerz).

63. Hypsirhina chinensis. Siam, China, Hainan, Hongkong
{Steindachner ').

64. HoMALOPSis BUCCATA. Bengal ?, Burma, Indo-China, Malay
Peninsula, Sumatiu, Borneo, Java, Hongkong {licdloto. <k City
Hall Mus.). — I examined the specimen in the City Hall Museum
which is labelled Hongkong.

65. DiPSADOMORPHUS KRAEPELixi : Boiga JiraepeXini Stejneger
(Proc. Biol. Soc. Wash. 1901, xv. p. 15). Formosa. — I found one
old specimen in the City Hall Museum labelled Formosa. This
specimen almost exactly accords with Stejneger's description, the
only points of diflterence being :^ — Temporals are six on one side.
Labials ten on one side. Anterioi- chin-shields in contact with
five lower labials on both sides. Yentials 244. Subcaudals 140.

66. DiPSADOMORPHUS MULTiMACULATTTS. Burma, Indo-China,
Malay Peninsula, Sumatra, Java, Celebes, S. China, Hongkong. —
A common snake in the south of China,. I obtained six from
Hongkong Island, one from Stonecutter's Island in that harbour,
and one from the mainland opposite Hongkong. There are
several specimens in the City Hall Museum, and I saw others in
spirit and in captivity belonging to Mr. Armstrong.

67. PsAMMODY^TASTES PULVERULEXTUS. E. Himalayas, Khasi
and Assam Hills, Burma, Indo-China, Malay Peninsula and
Archipelago, Formosa.

68. Chrysopelea orxata. Ceylon, Hills of S. India, Bengal,
Assam, Burma,, Indo-China, Malay Peninsula and Archipelago,
S. China, Hongkong ? ' — I found one specimen in the Siccawei

1 Reise der Novara, Kept., Wieii, 1869, p. 68.

- The specimen referred to by Boettger (Mat. lieryi. Fauna von China, p. 142)
is no longer in the City Hall Museum.

1903.] SXAKES OF CniXA AND JAPAX. 95

IVIusenm and one in the tSlifinghai Musenin. Suliciii'lals 1J4 in
one specimen.

69. AiPYSURUS AXxrLAxrs. Loyalty Islands, Loo Choos, 8ca,s
around Formosa {iS'iejneger)\

70. Hydrus platurus. Oliok, lied Sea, Indian Ocean, Straits
of Malacca, Tropical and Snl)tro]iical Pacific fi'om Loo Choo to
Australia and New Zealand, Malay Arcliipelago to Central
America. — I saw two specimens in the City Hall Museum fi'om
Hongkong and Formosa, and one in the Shanghai Museum. Also
two or three specimens in Mr. Owston's collection in Yokohama,
obtaineil t'l-om the Japanese shoi-es or the Loo Choos. All })elong to
variety E of the British Museum Catalogue of Snakes. Fi'ontal :
length less than distance to end of snout in one specimen, less than
parietals in two. Labials nine on one side in two specimens ; none
bordering the eye on one side in two sjjecimens. Anal tetrafid.

7L AcALYP'iDriiis PEROXii, Western Tropical Pacific.

72. HYDiiornis fasciatus. Coasts of India to China and Xew
Guinea.

73. Hydrophis GRACILIS, Coasts from Persia to China.
Hainan [Iler.z).

74. Hydrophis melaxocephalus. Indian Ocean, Pescadoi'es,
{Stejneger)", Loo Choos.

75. Hydrophis obscurus. Bay of Bengal to China. Canton
{Peters) \

76. DiSTiRA STOKESii. Mekran Coast to Chinese Sea, and Korth
Coast of Australia.

77. DiSTiRA ORXATA. Coasts of Asia from mouth of Persian
Gulf to Loo Choos, Kew Guinea, and N. Australia. — Tliis snake
evidently is connnon in the Loo Choos, as I saw more than twenty
specimens from that region in Mr. Owston's collection.

78. DiSTiRA SUBCIXCTA. Indian Ocean, Japan, Loo Choos
{Siejueyer) '.

' Sti'jiK'.troi- (.Jnnrn. Sc. (V)ll. Tokyo, xii. 18!)S-lfl()0, p. 223) <k'scrilK>s as !i uow
spet'ies uiultT i\w name of l<l>ui/docepIialus ijimce certain specimens, tlie ilesci-iption of
wliich I have consulted ]\Ir. Boulen.uer aliout ; and he is of ojiininn lliat these will he
found to \w ^\vm\\vu^ o\' A i fiiisHnni ail inihi f iig. The enlar<red vertehralsand ]n-esenco
of four iirn'trontals, found in these s]iecinu'ns, were also present in some specimens
I examined lielon.uin.n- to .Air. Owston in Yokohama, which [ re.ii'arded at the time as
anniilatnst. An examination of the specimens in the British Museum shows that a
certain slijrht cnlar;;cniunt of the vertehrals is present at some sj'ots, hut no specimen
has four iira'frontals. '{"he wci'^ht of .such an opinion compels me to modify the views
J had formed on the suhject.

- .J. Coll. .laiian. xii. p. 221 CMicroccplinlophis mihiiiorc[>h(tI ti!<).

» Monatsh. Herlin. Akad. 1872. p. 8o9 {H. dhttlcmn).

' Steineser (I'roc. Hiol. Soc. Wash. xiv. ]>. 1!)1. l'.)01) deseriltes certain specimens
oli'tained from the I'iu Kin Seas as Dixlim i>n<-)ii(i/is, which he considiM-s a new
si)ecie^. Kmm hi- ili-,-ri|it ion I am oti.pininn that these helonj;- to Jj. suhcinrfd.

96 CAPT. F. WALL ON THE [Feb, 3,

79. DiSTiRA BRUGMAXsii. Persian Gulf to Chinese Sea.
Hainan [jBoettg.) \

80. DiSTiRA CYANOCiNCTA. Persian Gulf to Chinese Seas,
Japan, Papuasia. — I saw one specimen of this snake in the City
Hall Museum labelled Hydrus viajor from Hongkong, and one in
the Shanghai Museum. One conforms to type A, and the other
to type C of the British Museum Catalogue.

81. DiSTiRA viPERiNA. Persian Gulf to Chinese Sea.

82. Enhydris hardwickii. Bay of Bengal to Chinese Sea and
Coast of New Guinea. — There are four specimens in the City Hall
Museum : two, labelled Hydrus major, are from Manila, and two
from Bangkok. There is one specimen in the Shanghai Museum
with very markedly spinose tubei'cles on the median six ventral
rows of scales. These rows are also enlarg'ed. Labials eight in
two specimens, in one on both sides and in the other on one side
only.

83. Platurus laticaudatus ^. Bay of Bengal to Chinese Sea,
Loo Choos, New Guinea, and Western South Pacific Ocean.
There are three specimens in the City Hall Mviseum labelled
Formosa, and I saw several specimens in Mr. Owston's collection
obtained from the Loo Choo Islands. Ventrals 232-246. Sub-
caudals 32-46.

84. Platurus colubrinus. Bay of Bengal to Chinese Sea,
Western South Pacific Ocean. — I examined two specimens, both
in the City Hall Museum, one from Penang and the other of
uncertain habitat.

85. BuNGARUS FASCIATUS. Southern Asia from Bengal to
China, Hongkong {City Hall Mus.). — I obtained one specimen
from the mainland opposite Hongkong, and I saw two others in
the City Hall Museum, one from Hongkong and the other from
the mainland opposite.

86. BuNGARUS CAXDiDus. Southern Asia from India to China,
Hainan, Hongkong {City Hall Mus.), Formosa. — I saw two speci-
mens in the Shanghai Museum, one in the Siccawei Collection,
and one in the possession of Mr. Styan in Shanghai. There are
four specimens in the City Hall Museum, one from Hainan and
three from Hongkong. All these specimens are of variety B of
the British Museum Catalogue, viz. midticincta. Postoculars :
the normal lower shield confluent with the fourth labial on both
sides in one specimen. Labials six, with the second and thh-d

1 Under the name of Sijdro^Jhis cyanncinctus Boettger (Mat. herp. Faun, von
China, 1888, p. 88) describes three specimens of what I consider Dis^ira hriigmansii ,
collected b}' Herz in Hainan.

^ The species described by Boulenger (Cat. Snulces Brit. Mus. iii. p. 309) as Platurus
muelJeri I do not believe to be valid, and I think the specimens on which it is
based will prove to be F. laticaudatus. The only definite point he mentions to
characterize it is the presence of a median ventral keel in the posterior half of the
body, and I have found this peculiarity in at least three specimens of what I consider
uudoubted P. laticaudatus.

1903.] • SXAKES OF CHINA AND JAPAN. 97

touching the eye on one side in one specimen, and occasioned by a
confluence between tlie first and second labials. Anterior chin-
shields in contact with four infralabials on both sides in tliree out
of four specimens examined.

87. ISTaia tripudians. Southern Asia from Ti-anscaspia to
China, Hainan, Hongkong [Jlalloio., Steindach., City Hall Mus.),
Formosa, Chusan, Philippines. — In the Siccawei Museum 1 saw
seven specimens, and in the Shanghai Museum three. There are
fifteen in the City Hall Museum, of which eleven are from Hong-
kong, two from Hainan, and two from China. Of the ten
specimens I examined three have scales in twenty-one rows in
the middle of the body, and three in nineteen. In the remainder
I have failed to record the number. Nine have more or less
distinct (some very well defined) buft" or pale yellowish cross-bands
dorsally. These numbered from 13-21 on the body, and 5-8 on the
tail. They involve one or two rows of scales along the length of
the snake, and are most conspicuous in the posterior third or so
of the body. The intervals involve fiom ten to twelve scales, and
are sometimes sepai'ated from cross-bars by a blackish line. Hood
marked with modified black ocellus. Belly yellow, with one or two
plumbeous bands ventrally, or mottled to a variable extent with
black. In one specimen the whole belly uniform black. One
specimen is evidently a Sputatiix. This I found in the Siccawei
Collection. Scales over hood 21, mid-body 17. Ventrals 177.
Subcaudals 45. No hood-marks. Nearly unifoi-m black dorsally,
with no suspicion of cross-bars. One very broad plumbeous ventral
band involves from the 9th to the 47th shield, and then breaks up
into a mottling and disappears.

88. Naia bungarus. Southern Asia from India to S. China,
Hongkong {City Hall Mus.), Philippines. — A newly-hatched ex-
ample is preserved in the City Hall Museum, habitat Hongkong.
I found the head of a large specimen in the Shanghai Museum ;
and whilst I was in Hongkong a gentleman encountered one ou
the mainland opposite, which he killed and bi'ought home. I
examined and identified it, and estimated it at between seven and
eight feet in length.

89. Callophis macclellandii. Nepal, Sikkim, Assam, Burma,
S. China, Formosa,

A M B L Y C E P 11 A L I D .«.

90. Amblycephalus moellendorffii. Tenasserim, Siam, Cochin
China, S. China, Hongkong, Hainan. — This is a common snake
in the island of Hongkong. I collected ten myself, all of which
were found in low jungle on the slopes of the Peak. I saw three
specimens in the City Hall Museum. Internasals form a suture
with the loreal in all specimens. Postoculars and suboculars :
usually one long semilunar shield extends from the supraoculars
behind, skirting the eye to a point about halfway up its anterior
aspect. Sometimes this scale is divided so as to form a small

Proc. Zool. Soc— 1903, Vol. I. No. YII. 7

98 CAPT. p. WALL OX THE • [Feb. 3,

postocular above this shield. Temporals : the antero-superior
extends back as far as the extreme border of the parietals or
further. Labials : six in two specimens on both sides. Sub-
caudals 36-50. First infralabials meet in four, fail to meet in
one, not recorded in eight.

91. PsEUDOPAREAS VAGUS. Hongkong (Jan).

YiPERIDJE.

92. Ancistrodon acutus. China. — Called by the Chinese
" Oo-woo-shay," or five-pace snake, owing to the reputed rapidly
fatal eifects of its bite. I saw one specimen in the City Hall
Museum labelled A. blomhoffii, habitat Szechuen. There are also
five specimens in the Sicca wei Collection. Two are labelled Ou-
Yuen\ and one of these dated 1882 (i. e. six years before Giinther
first described it). One is labelled Chowtong" 1883, and another
Kien-te^ 1882. The fifth consists of a dried skin. In talking
to the late P^re Heude about this creature, he told me it is not
uncommon in hilly parts of the Yangtse Yalley, and is feared by
the natives more than a leopard. He narrated an experience of
one of the Jesuit Fathers who heard frantic cries for help issuing
from some dense jungle he was riding through. On dismounting,
and proceeding to the spot, he found a Chinaman being pursued
by one of the snakes, which he attacked and killed. Rostral : as
far as I could ascertain there is only one shield beneath the dermal
appendage, which, however, is frequently bent over, causing a
fuiTow which makes the shield appear divided into two. Labials :
six in one specimen on both sides, and in another on one side.
Subcaudals 56 in one specimen, with Ist-llth entire and rest
divided ; 53 in another, with lst-9th entire and rest divided ; 58
in another, with lst-28th entire and rest divided ; 57 in another,
with lst-12th entire and rest divided.

93. Ancistrddon blomhoffii. E. Siberia, Mongolia, China,
Hainan, Formosa, Japan, Loo Choos, Siam. — Mollendoi-flf says it
is called by the Chinese " Fei-shang-ts'ao," snake which flits on
the grass, and " Ch'i-ts'un-tsze," or seven-inch snake. It is very
common in the Yangtse Yalley. There are at least three specimens
in the City Hall Museum from Japan. There are five in the
Shanghai Museum, and a very large number in the Siccawei
Collection. In Mr. Owston's collection I saw several obtained
from Japan and the Loo Choos. I obtained five myself about
Shanghai in our camp. Two of these, which I found close together
on October 16th, 1901, appeared to be newly hatched. On the
4th October, 1901, my servant informed me that there were several
snakes lying together dead close to the camp. I investigated the
matter, and found an adult of this species with the remains of

1 Perhaps Tao-yuan in Province Hunau ?
" Perhaps Chao-tung- in North Yunnan ?
5 In Province An-Hui.

1903.] SXAKES OF CIIIXA AXD JAPAX. 99

twelve ynimg PCiittered ai'ouud her. In spite of every endeavour,
I failed to tr:ice the author of this butchery. It is .«jid luck tliat
such oppoi'tunities as these for oliservino- hahits I'ai'ely fall to the
lot of tliose who specially seek them. Residents ahout Shanghai
occasionally lose dogs under circumstances suggesting a poi.sonoua
bite, and I think these fatalities are frequently due to this little
snake. One specimen I cjiptvu-ed had the penis hifid on both sides.
I kept young on two or thi-ee occasions, but could not find suitable
food with which to tempt them, and finally I'esorted to feeding
them with boluses of raw meat, a treatment they displayed their
distaste for l)y repeatedly disgorging the morsels some time later.
I observed that, when molested and excited, all my specimens
vibrated the tail, but I could never get one actually to sti'ike ;
however all the specimens I had alive were veiy small. Labials
seven, the third only touching the eye ; eight in one specimen on
both sides.

94. Lachesis jerdoxii. Ivhasi Hills, Assam, Thibet ; Upper
Yangtse, China.

95. Lachesis mucrosquamatus. Kaga Hills, Assam ; Formosa.

96. Lachesis cramixeus. South-eastern Asia, China, Hainan
{Herz (I- Cit]/ Hall Mas.), Hongkong, Formosa. — This is a veiy
common snake in Hongkong, whci'e I procui'ed three specimens.
There are also twelve in the City Hall Museum, obtained locally.
In one of these the skin of the neck had been perforated and
rent by the beak of a bird which it is seen in the act of devouring.
The distension of the mouth and neck is enormous, vso that the bird
appears as if it were being thrust through the rent instead of into
the gullet. Thei'e is only one specimen in the Shanghai Museum,
and only one in the Siccawei Collection. The lesidents of Hong-
kong call this the Bamboo-snake, which is a very good name for it
since it is almost always to be found in the foliage of bamboo
vegetation. I am told too that the Anglo-Indians in Assam have
also given this name to the snake.

Part II. — List of Japaxese axd Loo Choo Islaxds Ophidia.

C O L U B R I D --E.

1. Tropidoxotus vibakari. Japan, Formosa, Manchuria. — I
found one in the City Hall Museum from Nagasaki, and saw six
others in Mr. Owston's collection, all from IVlount Fuji. Ventrals
152 in one specimen. Subcaudals 80 in one specimen; second to
fifth entire in one specimen.

2. Tropidoxotus tigrixus. Siam, Cochin China, China, Hainan,
Hongkong ?, Manchuria. — In my opinion this is the commonest
snake in Japan. I frequently met with it, and on one occa'^ion
captured five in a couple of hours. It is called by the Japanese
"yamakagashi," " uwabami," '-orochi," "'ja," all of which terms, I

7*

100 CAPT. F. WALL ON THE [Feb. 3,

am informed, signify " the largest variety " ; also " tora-no-kuchi-
nawa," or "tiger-snake"; also " miza kuchinawa," or "water-
snake"; also "atsuki kuchinawa," or "thick brown snake." (See
also No. 12 of Chinese list.)

3. Tropidonotus prteri. Loo Choos. — I saw one specimen in
Mr. Owston's collection from ISTawa in the Loo Choos.

4. AcHALiNUS SPINALIS. China, Japan. (See No. 26 of Chinese
list.)

5. DiNODON RUFOZONATUS. China, Hainan, Formosa, Chiisan,
Corea, Tsn Shima, Japan, Loo Choos. (See No. 30 of Chinese list.)

6. DiNODON SEMICARINATUS. LoO CllOOS.

7. DiNODON JAPONicus. Japan. — I found one specimen on the
Southern Island (Kiu Siu), and saw another in Mr. Owston's
collection. Yentrals 209. Subcaudals 75 ? and 77.

8. DiNODON TESSELLATUS. Japan,

9. Coluber dione. S. Russia, Transcaucasia, Temperate Asia,
China, Hainan, Japan (Blgr.). (See No. 39 of Chinese list.)

10. Coluber schrenckii. Amoorland, Corea, N. Japan.

11. Coluber conspicillatus. Japan, Corea. — I saw one
specimen in the City Hall Museum labelled Nagasaki and another
in Mr. Armstrong's collection captured in Japan. Called by
Japanese "jimuguri" ("earth borer") and " kawara kuchinawa"
(" dry-river-bed snake ").

12. Coluber climacophorus. Japan. — A very common snake
in Japan, where it is called "aodaisho," "nezumi tori" (or " i-at-
catcher"), "sato meguri" (or "village idler"), and " mugiwara hebi."
I frequently encountered it. I saw and captured one reclining
on a stone parapet in one of the Nikko temples. I discovered
another in thick jungle in the act of swallowing a half-grown
leveret, ti-uly an enormous meal, since the mammal was more
than twice the weight and girth of the reptile. I thought on
all occasions it was far less wary and active than other nearly
allied snakes of a similar size with which I am familiar, such, for
instance, as Zamenis 7)111008118 and Z. korros, Coluber radiatus,
Zaocys dhumnades, &c. Temporals, in one specimen, three on
both sides.

13. Coluber quadrivirgatus. Japan, Corea. — This species is
almost if not quite as common in Japan as Tro^ndonottis tigrimis.
I scarcely went a day in the coiintry without seeing one, and
often I saw three or four. It is called by the Japanese " shima
hebi" (or "striped variety"), "karasu hebi" (or "black snake"),
" sukuro hebi" (or " medium black snake "), " mugiwara hebi " (or
" wheat straw snake"), and " kiiro kuchinawa." Subocular absent
in one specimen on one side. Labials eight, with third, fourth,
and fifth touching the eye on one side in one specimen.

1903.] SNAKES OF CHINA AND JAPAX. 101

14. Coluber schmackeri. Loo Choos.

15. Ablabes semicarixatus. Japan, Loo Choos.

16. Ablabes hermin.e. Loo Choos. — I saw two specimens in
Mr. Owston's collection from the Loo Choos. Ventrals 162-163.
Subcaudals 47-51.

17. Calamaria pavimextata. Burma, Siam, Cochin Cliina,
Malay Peninsula, Java, S. China (Mull.), Riu Kiu Archipelago
(Stejneger). (See No. 56 of Chinese List and footnote.)

18. Hydrus platurus. Coasts of Asia from Red Sea to Loo
Choos, Australia, New Zealand, Malay Archipelago to Central
America. — Called " umi hebi" or "sea snake" by Japanese. (See
No. 70 in Chinese list.)

19. Hydrophis melaxocephalus. Indian Ocean, Pescadores
{Stejneger), Loo Choos. — This is evidently a common snake about
the Loo Choos, since I saw upwards of twenty specimens in
IMr. Owston's collection from this locality. Postoculars one or two.
Temporals : one anterior usually, sometimes two. Scales two
head-lengths from snout 23-27, mid-body 31-35. Yeutrals 305-
343. Anal tetrafid. (See No. 74 of Chinese list.)

20. DiSTiRA cyaxocixcta. Persian Gulf to Chinese Seas,
Japan, Papuasia. (See No. 80 of Chinese list.)

21. DiSTiRA ORXATA. Coasts of Asia from mouth of Persian
Gulf to Loo Choos, New Guinea, Northern Australia. (See
No. 77 of Chinese list.)

22. DiSTiRA subcixcta. Indian Ocean, Japan, Loo Choos
{Stejneger). (See No. 78 of Chinese list and footnote.)

23. Aipysurus axxulatus. Loyalty Islands, Loo Choos, Seas
around Formosa {Stejneger). (See No. 67 of Chinese list and
footnote.)

24. Platurus laticaudatus. Bay of Bengal to Chinese Sea,
Loo Choos, New Guinea, Western South Pacific Ocean. (See
No. 83 of Chinese list.)

25. Platurus colubrixus. Bay of Bengal to Chinese Sea,
Western South Pacific Ocean. (See No. 84 of Chinese list.)

26. Platurus schistorhyxchus. Western Pacific Ocean, Loo
Choos.

27. Hemibuxgarus japoxicus. Loo Choos, Japan ?

Viperid^.

28. AxcisTRODOx ixxESMEDius. Japan, Mongolia, Ea^ern
Siberia, Central Asia.

102 ME. LYDEKKER ON AN ASIATIC WILD SHEEP. [Feb.-S,

29. Ancistrodon blomhoffii. Eastern Siberia, Mongolia,
China, Hainan, Formosa, Japan, Loo Choos, Siam. — Called by
the Japanese " mamushi," " hami," " knchibami," "hiraguchi."
All synonymous terms for the " beautifvilly marked variety."

30. Lachesis okinavensis. Loo Choos. — Called by the Japanese
" habu."

3L Lachesis flavoviridis. Loo Choos. — Also called "habu"
by the Japanese.

32. Lachesis mucrosquamatus. Naga Hills, Assam ; Formosa.

33. Lachesis luteus. Loo Choos.

3. Note on the Wild Sheep of the Kopet-Dagh.
By R. Lydekker.

[Eeceived December 4, 1903.]
(Text-figure 10.)

Tlirough the generosity of Mr. St. George Littledale the collec-
tion of the British Museum has recently been eniiched by a very
fine skull, with the horns (text-fig. 10, p. 103), of a ram of the Wild
Sheep of the Kopet-Dagh i-ange, which forms the boundary between
Turkestan and Noi-thern Pei-sia to the eastward of the Caucasus.
Mr. Littledale also brought home the skin of the same animal,
bvit it unfortunately was so badly injured by vermin that it had
to be destroyed. The skull is the finest of the series obtained
during the trip.

The Kopet-Dagh sheep was named Ovis arkal in 1857 by
Blasius, and is evidently allied to the Urial, with the Punjab
race of which I have indeed proposed to identify it \ At that
time I had, however, never seen an adult skull ; and Mr. Little-
dale's specimen indicates the right of this sheep to rank as a
distinct race of Uriah It wall be remembered that the Punjab
race of the Urial {0. vignei cycloceros), as exemplified at any rate
by specimens from Peshawer and Afghanistan in the British
Museum, difiers from the typical 0. vignei of Astor and Ladak
by the much greater prominence of the two front angles of the
horns, which are often raised into nodose beads, between which
the front surface of the horn is depressed and carries bold and
widely separated ti'ansverse ridges.

In the Kopet-Dagh Urial this prominence of the front angles
of the horns is still more pronounced, though the beading is
somewhat less conspicuous. Moreover, the fi'ont surface of the
horn is unusually broad and flattened, with the transverse

1 ' Wild Oxen, Sheep, and Goats,' p. 173.

1003.]

MU. LYDKKKEH ON AX ASIATIC WILD SlIEEr.

103

♦wi'inkles very low and indistinct. The It'ngth of the horn is
33 inches along the inner front angle, with a basal circnniference
of 11 inches, a basal width of 3 inches, and a ba.sal depth of
4 inches.

The last two dimensions are considerably greater than in a
skull of the Kelat Urial measured by Mr. Hume, in which the
length along the curve is 35| inches.

Text-tiij. 10.

Front view of skull and lionis of adult rain of the Kopet-Dagh Urial.
5 nat. size.

The Kopet Dagh Urial decidedly appears to be a distinct form,
connected with the typical Ocis vignei by means of the Punjab
race of that species. On these grounds I regard it as a local race
I'ather than a species ; its name will accordingly be 0. vignei ai-kal
(or perhaps arcal). The suggestion of M. Dauvergne \ that this
sheep is identical with the Kelat Urial, is not borne out by a
comparison of the present specimen with a skull of that form in
the British Museum, in which the angles of the horns are much
rounded ofi\ This leads me to think that the Kelat Urial
{0. vignei blanfordi) is, after all, distinct from the Punjab
animal.

1 See p. 131 of my book on 'Game of Europe, N.W. Asia, and America.'

104 MR. P. W. BASSETT-SMITH OlST [Feb. 3,

4. On new Parasitic Copepoda from Zanzibar and East Africa,
collected by Mr. Cyril Crossland, B.A., B.Sc. By Staff-
Surgeon P. W. Bassett-Smith, R.N., F.Z.S.

[Received December 4, 1902.]
(Text-figures 11 & 12.)

Mr. Cyril Crossland, in his recent examination of the marine
fauna of Zanzibar and British East Afi'ica, obtained several
specimens of parasitic and semiparasitic Copepods, three of which
he has been kind enough to allow me to examine.

These curiously deformed and often grotesquely-shaped animals
are frequently found attached to the gills, &e., or to the surface of
fish and other marine animals.

A large number from the former which ai-e now described from
a variety of different hosts, and from wide geographical areas,
I enumerated in Proc. Zool. Soc. 1899, p. 438 ; to this paper
I have appended a list of addenda, which I have drawn up with
the kind assistance of Mr. E. Bergrotti, of Tammerfers, and
others.

Not only fish bvxt a number of other marine animals are un-
doubtedly infested with these parasites, though at present little
information concerning them has been collected ; the specimens of
Mr. Crossland are therefore particularly interesting.

Gerstacker, in Bronn's ' KJass. unci Ordn. des Thier- Reich s,'
1866-79, Crustacea, vol. v. Copepoda, p. 774, mentions five genera
found on Nudi branch Mollusca : Doridicola Lyd., Eolidicola Sars,
belonging to the family Ergasilidae ; Ai-totrogus Boeck, to the family
Ascomyzontidse ; and Siolanchnoiroinis Hanc. and Ismailia Bergh,
to the family Chrondi'acanthidas. Also nine genera fi'om various
Vermes, p. 773.

Of the three specimens of Mr. Crossland, two were taken from
the kidneys of species of Pleurobranchids (not determined) and
one from the skin of a Sipunculid (^Aspidosiphon).

As they were only single specimens it was impossible to dissect
them, and therefore the descriptions are necessarily incomplete.
The first two evidently belong to the family Chondracanthidre,
but do not fall in with the descriptions of any known genus ; they
appear to be most nearly related to the genvis Splanchnotropus of
Hancock (Trans. Linn. Soc. vol. xxiv. pp. 51, 55), two species of
Avhich he describes, *S'. gracilis and S. brevipes, taken from Nudi-
branchs ; the present specimens differ from them, however, in the
complete absence of antennae and articulate limbs, and in having
the external ovaries elongated and the eggs arranged in single
series.

I have therefore provisionally placed them in a new genus,
Cho7ulroca7-pus, following closely after S2^lconchnotro2n(,s Hanc. and
IHocus Fabr.

Chondrocarpus, gen. nov.

$ . Cephalothorax coriaceous, elongated, with four lateral short

1903.

NE\Y PARASITIC COrEFODS.

105

lobe-like processes on either side ; abdomen one-fourth length of
whole, biiU'tieulate, tapering ; no distinct antennae or thoracic
limbs ; mouth placed on under surface, with minute maxillte ;
external ovaries as elongated filiform sacs containing ova in a
single series,
d. Pigmy.

Chondrocarpus reticulosus, sp. n. (Text-fig. 11, A-G.)

llah. Zi;\\\z\\rAV : from large Pleurobi'anchid,

Length 12 mm., breailtli 4 nnn.

2 . Oephalothorax indistinctly segmented, elongated, tapering,
dorsally convex, laterally produced into 4 rounded truncated
processes ; the 4th pair being most widely separated and the
smallest ; the whole having a peculiar I'eticulate appearance from
the network of ramifying tubules; anterior extremity rounded;
no visible antenna; ; mouth as a papilla placed between the 1st
pair of processes ; upper lip triangular ; oidy one pair of maxillae
and mandibles (?) could be made out, each terminating in minute
claws. No thoi-acic limbs; there is a short genital segment, or^
ring, from which spring the long filiform ovaries. Abdomen in-
distinctly biarticulate, tapering, without caudal plates or seta\

<S . Pigmy. One was seen attached to the last abdominal
segment but was partially hidden, the bifid ai'ticulate caudal
extremity only being visible.

Text-fig. 11.

Chondrocarpus reticulosus ? , pren. & sp. ii. A. Dorsal surface. B. Lateral view.
C. First segment and mouth-organs. D. One of tlic lobes sliowing reticu-
late appearance. E. Second maxilliped. 1<\ Abdomen showing fixed j.
G. Posterior extremity of (J. H. C/iondrocarjius sp. $ : dor»al surface
(specimen incomplete).

CnoxDRocARPus sp. (Text-fig. 11, II.)

A specimen of a second species of this genus was taken from a

106 MR. P. W. BASSETT-SMITH ON [Feb. 3,

Pleurobranchid ; it was much broken both at the anterior and
posterior extremities. It differs from G. reticidosiijS in not having
the peculiar reticulate appearance and in having a pair of lateral
lobes on either side of the genital segment.

The third specimen also appears to be new, and belongs to the
family Dichelesthiidfe, the animal resembling most nearly the
genus Enterocola of Van Beneden (Bull. Acad. Roy. de Belg.
torn. ix. 2nd ser. p. 151), found by him in the respiratory cavity of
" Aphidium Jicus," than any other form I have been able to find
recorded.

Mr. Crosstand's specimen appears to be much more degenerate
from its parasitic habits ; the ai'ticulate limbs are excessively
small, difficult to make out, and the ova are carried in long spiral
thread-like processes in a single series as in the Caligidse, and
not in dilated sacs. Unfortunately there was only this single
female specimen for examination. I would pi'ovisionally ci-eate
for it a new genus " Ventriculina," giving the specific name of
" crosslandi" in recognition of the collector.

Ventriculina, gen. nov.

Head small, rounded ; neck indistinct ; 3 thoracic segments,
the first amalgamated with the head ; genital segments lobed,
equal in breadth with the thoracic ; no dorsal plates ; abdomen
short, biarticulate ; external ovaries spiral, ova uniserial ; first
antennse 4-jointed, simple, second antennae 3-jointed ; maxillipeds
very small ,

Three pairs of minvite thoi'acic limbs, first biramose, second and
third uniramose.

Ventriculina crosslandi, sp. n. (Text-fig. 12, p. 107.)

Hah. Zanzibar : from a Sipunculid.

Total length 4 mm. Colour white.

Head small, rounded in front, broadest behind, from under
which pi'oject the anterior antennae. Thoracic segments three,
the first united with the head, the genital segments trilobed ; the
whole forming an oblong body without lamellar plates, showing
five distinct rounded lateral lobes, the last pair being slightly
wider and more acute. Dorsal surface convex, marked by five
distinct grooves showing the position of the union of the segment^.

Abdomen biarticidate, narrow, one-sixth the total length, the
last joint terminating in two small caudal plates provided with a
marginal fringe of short bi'istles. External ovaries long, spiral,
spi'inging fi'om papillse at the angle of the genital segments and
abdomen. Ova large, arranged uniserially as in the Caligidse.

Anterior antennae 4-jointed, non-se''jOse, the first joint being
the longest and broadest, rising from the underside of the head
just in front of the rnouth ; second, third, and fourth joints
progressively decreasing in size.

1903.]

NEW PARASITIC COPEPOES.

107

Posterior antennre S-joiuted, rising just in front of the upper
lip ; at the distal end and anterior border of the first and second
joints are two short setje, the third terminating in two bristles,
the anterior being very long.

Mouth and appendages placed rather far back ; the labrum is
tiiangular, projecting backwards ; labium simple, I'ounded. I was
able to make out only 2 pairs of maxillipeds ; the first very small,
biarticulate, terminating in tw^o short hairs : the second uncinate,
with large globose basal joint, to wdiich was articulated a sharp
curved claw.

Text-fig. 12.

Vciitriculhia crosslandi $, gen. & sp. n. A. Ventral surface, X 10. B. Dorsal
surface. C. Ventral surtacc much enlarged, showing articulate appendages.
D. Posterior antenna;.

Only three pau-s of thoracic limbs present : the first rising f i-om
the posterior under sui-face of the cephalic segment, minute,
biramose, each i-amus terminating in a single bi'istle ; second and
thii'd paii's uniiumose, made up of two ai'ticulations, the distal
tei-minating in two small bristles.

(S not known.

Addenda to Systematic Enumeration of Sjxicies of Parasitic
Copepoda found on Fish (Proc. Zool. Soc. 1891), p. 438).

Ergasilid^.

1. Eucanthns viarchesettii Yalle, Atti Mus. Civ. Trieste, vii.

p. 245 (1885). On Jlotella tricirrata.

2. Ergasilas centrarchidaricm Wright, Proc. Canad. Inst. (2) i.

p. 243 (1883).

3. Eryasilus biuncinatus Gadd. Meddelanden af Societas pro

Fauna et Flora Fennica, xxvii. pp. 181 182 (1901). On
Gastrostens acnleatns.

108 ON NEW PARASITIC COPEFODS. [Feb. 3,

4. Bomolochus onosi T. Scott, 20th Ann. Rep. Fish. Board of

Scotland, p. 289, pi. xiii. figs. 19-22 (1902). On Onos
musiehts and O710S cimbrius, Firth of Forth.

5. Bomolochus zeugopteri T. Scott, loc. cit. p. 290, pi. xiii. figs. 23-25.

On Zeugojiterus punctatus.

CALIGIDiE.

P. Caligus pacificus Gissler, Amer. ISTat. p. 886 (1883). On
Salmo.

7. Caligus labracis T. Scott, I. c. p. 291, pi. xiii. figs. 26-29. On

Lahriis mixius and L. 'macidatus.

8. Anchicaligtis nautili Stebbing, Willey's Zool. Res. pt. v. pp. 667-

670, pi. Ixxi.

9. Dinematura mtisteli Icevis Hesse, Rev. Sci. Nat. Montpellier,

(2)ii. pp. 6, 11 (1880).

10. Cecrops acanthice vulgaris Hesse, Ann. Sci. ISTat. (6) xv. 3,

p. 26 (1883).

11. Pandarus carcharii glaucus Hesse, I. c. p. 18.

12. ,, musteli Icevis Hesse, I. c. p. 23.

13. ,, spinacis acanthice Hesse, I. c. p. 10.

14. ,, unicolor Hesse, I. c. p. 20. On Gcdeus vulgaris.

DlCHELESTHIID^.

15. Lernanthropus polyneini Rich. Zool. Anz. iv. 1881, p. 505.

On Polynevius.

16. Lernanthropus tetradactylus (probably L. trifolicUus B.-S.

1898).

17. Lernanthropus micro2)terygis Rich. Atti Soc. Tosc. Sci. Nat.

iv. p. 82 (1884). On Micropteryx dumerili.

18. Lernanthropus tylosuri Rich. I. c. p. 83. On Tylosurus im-

pericdis.

19. Kroyeria (Lonchidium) gcdii vidgaris Hesse, Ann. Sci. Nat.

(6) xvi. 3, p. 2 (1883).

20. Clavella cluthce T. Scott, l. c. p. 292, pi. xii. figs. 26-31, On

Ctenolabrus rupestris.

21. Pagodina (Nemesis) charcharice glauci Hesse, I. c. p. 13.

22. Eudactylina carcharice glauci Hesse, I. c. p. 11.

23. ,, musteli Icevis Hesse, I. c. p. 8.

24. „ squatina ccngeli Hesse, I. c. p. 5.

25. „ similis T. Scott, I. c. p. 295, pi. xii. figs. 1-19.
On Raia radiata.

26. Eudactylina acanthii T. Scott, I. c. p. 296, pi. xiii. figs. 1-9,

On Squalus acanthias.

27. Bassettia congri Stebbing, Willey's Zool. Res. pt. v. pp. 671,

672, pi. Ixx.

PhILICHTHYIDjE.

28. Philichthys ficdolce Rich. Zool. Anz. iii. p. 69 (1880). On

Stromateus jialola.

29. Philichthys doderleini Rich. Zool. Anz. vi, p. 558 (1883). On

Labrus turdus.

1903.] ox THE ORIGIN'Ar. HOME OF THE TIOER. 109

Lerx.eid.e.

30. Lerncea ahyssicola Brady, Chall. Rep. viii. p. 137. On Circdias

m'cinoscopus.

31. Lerncea minuta T. Scott, IStli Rep. Fish. Boai'd of Scotland,

p. 161, pi. vii. fig. 13 (1900). Oa Gobiiis miniUas.

32. Lermea lumpl T. Scott, 19th ditto, p. 128, pi. vii. fig. 12

(1901). On C)/clopterus lunipus.

33. Hfemohaphes amhigims T. Scott, 18tli ditto, p. 1G2, pi. vii,

fig. 15. On CalUonymus maculatas.

34. Peraderma petersi Rich. Zool. Anz. iv. 1881, p. 387. On

Goh'ms huGcatiis.

35. Peraderma hellottii Rich. Zool. Anz. v. 1882, p. 475. On

iScopelus henotti.

CHONDRACAXTniD^.

36. Chondracanthus hleeheri Rich. Zool. Anz. iv. p. 387 (1881).

On Chiliiim chlorurus.

37. Chondracanthus ninnii Rich. Zool. Anz. v. p. 504 (1882). On

Gohias.

38. Chondracanthus ornatus T. Scott, 20th Ann. Rep. Fish Board

of Scotland, p. 298, pi. xiii. fig. 34. On CalUonymus
maculatus.

Lerx.eopodid.e.

39. Achtheres sandrce Cladd. Med. af Soc. pro Fauna et Flora

Fennica, xxvii. (1901).

40. Lernceopoda extumescens Gadd. I. c. On Coregonus.

41. Tracheliastes gigas Rich. Zool. Anz. iv. 1881, p. 504.

42. Charopiims dubius T. Scott, 19th Ann. Rep. Fish Board of

Scotland, p. 130, pi. vii. fig. 15. On Paia circular is.

5. On tlie Original Home of the Tiger.
By Col. C. E. Stewart, G.B., C.M.G., C.I.E.i

[Received December 6, 1902.]

The ordinary idea of English people that the Tiger was origin-
ally an Indian animal, is, I believe, quite a mistake. After
careful enquiry, I have come to the conclusion that the Tiger is a
comparatively late intruder into India.

Firstly, after enquiry, I cnn discover no Sanscrit word for the
Tiger. If tigers had existed in India in the days when Sansci-it
was a spoken language, there would he a name in Sanscrit for it,
while there is only a modern Hindustani name. There is a
Sanscrit word for Lion, " Singha," which would point to the
fact that lions were certainly more common than tigers in time
long past. At present lions are not found in India, except a
very few, which are strictly preserved in Googerat, one extreme

' ComniuiiicateJ by Col. Hili, .James, F.Z.S.

110 ON THE ORIGINAL HOME OF THE TIGER. [Feb. 3,

corner of India, tliough I will allow that lions were probably
commoner tlian they are now in the olden time, though probably
never veiy numerous.

I remember, when I first went to India, nearly 50 years ago, a
lion being killed not very far to the southward of Allahabad, but
this was even then a rare occurrence. I have studied the
question of the habitat of lions and tigers in Persia, where I
resided for a good many years. Lions are found only in the
very south of Persia, near the Persian Gulf, and Arabia ; while
tigers are only seen in the very north of Persia, near the Russian
border, and especially near the Caspian Sea, on the north of
Persia, and they are moi-e numerous within British territory
than within the Persian boundary, and tigers are more common
in Southern Siberia than they are anywhere in Persia.

Tigers are more numerous in cold countries. They are plen-
tiful in Oorea, which has a severe winter climate, and still more
plentiful in the Island of Saghalien, belonging to Russia, and
further noi'th than Oorea, and which has almost an ai'ctic climate
in winter. The tiger is mentioned by Marco Polo in his travels, but
nowhere as an Indian animal, and I very much doubt whether
tigei'S were found in India at the time Marco Polo visited it.

In the Sanscrit woi'ks treating of the fighting between Rama
and Rawun, the Demon King of Ceylon, though many animals
are mentioned, such as beai-s, monkeys, and several others, I
have been unable to find any mention of the tiger ; and the tiger
is not found in the Island of Ceylon, though the leopard is ; nor
is the tiger found in the larger island of Borneo, which would
seem to point to its only inhabiting the islands of the Indian
Archipelago, which it could reach by swimming. Thus it would
seem that tigers did not exist in India before the time that Ceylon
was separated from India. Tigers did not exist in the island of
Singapore until about 1809, when apparently they swam over
from the mainland. Tigers are such good swimmers that they can
cross a considerable body of water. I do not think any allusion
to tigers in India can be found in the Greek historians. I should
feel much obliged if anyone could find me such a reference.

In the monuments of the Assyrian Kings, and of the Kings of
Persia, there are constant references to lion-hunts by those kings,
but never allusion to a tiger-hunt. Of course there is an existing
Persian word for tiger, but there is nothing to show that it is at
all ancient.

My own idea is that the tiger was originally a purely northern
animal, which has gradually extended southward. I fancy that
no allusion to a tiger in India can be traced to a period anterior
to the early Mahommedan conquerors of India. I should be
much obliged to anyone who will help me to clear up this
question. We English have so completely assumed the idea
that the tiger is an Indian animal, that we have called him the
Royal Bengal Tiger, though I firmly believe he is as much an
intruder from the noitli into Bengal as we are ourselves.

1903.] ox THE roruLATiox of the ixdiax elephant. Ill

6. On the Mode of Copuliitioii of the Indian Ele]»hant.
By H. Slade, Conservator of Forests, Maymyo, Burma.^

[Received December 15, 1902.]

Some few months ago I was, on several occasions, al)le to witness
tame Elephants in tlie act of copulation. This sight has been so
seldom witnessed by Europeans, and is so variously described l)y
Burmans, that these remarks, supported V)y a sei'ies of photographs
deposited with the Society, may be of interest and value.

As tuskers are usually i-eputed to be shy of copulating before
eye-witnesses, an account of the manner in which these photo-
graphs wei-e procured may be interesting.

I wjxs in camp at the time with one tusker and four female
Elephants, which were being used regulai-ly for ti'ansport pui'ijoses.
From the time the tusker was i-epoi-ted to be seeking the company
of the females he was never let loose to gi'aze, but was kept tied
up. I had pi'omised the Bui'mese mahouts libei'al i-ewai'ds if they
would assist me in procuiing some " snap-shots"; and one djiy last
Febi'uary one of them came to tell me the tuskei- was showing
undoubted excitement ; he was reported to have been tugging at
his chain and looking " nastily " at his keeper. There weie,
however, absolutely no signs of " must " and no exudation of fluid
fi'om the hole in the temple.

The tuskei' was said to have shown a prefei'ence for one of the
females that had calved about 12 months pi-eviously. So I had
her tethei-ed foi-e and aft in a small glade, and ei'ected my camera
quite in the open, about 10 yards off. Having focussed her and
got eveiything I'eady, the tusker was quietly ridden up behind
the female, and as the mahout slipped ofl", he slowly advanced
towai'ds hei'. However, this was evidently the WTong female, for
on the appi'oach of the tusker she showed most unmistaka1)le
signs of fear by pei'sistently screaming and straining at hei' chain.
The tusker was therefore secui'ed, and this female marched oft',
whilst one of the others was caught and tied up in her place.
The tusker was then again brought out, and immediately pi-oceeded
to mount the female.

I had many oppoi-tunities of witnessing the operation, and the
following is a general desci'iption : — The female, when in season,
remains perfectly still and quiet, merely signifying her sense of
the tusker's approach by moving hei- tail slightly to one side and
gently shifting hei- hind feet a few inches further apart. Tlie
tusker creeps up behind the female and begins to show signs of
sexual excitement. He then I'aises his head and lightly places
his tusks on the female's back, one on either side of her backbone,
with his trunk lying along the dorsal ridge and reaching to lier
shoulder. In this position he I'emains a length of time, varying
with his state of sexual excitement. From my oliservations 1 am
convinced that when he lisis uniestrieted access to females, and is

^ Coiniminicatcd Ii}- tlie Seckktaky.

-il2 ON THE COPULATION OF THE INDIAN ELEPHANT. [Feb. 3,

not therefore greatly excited, he may remain a very considerable
time in this position : this has probably given rise to the tales
that are told of the great length of time occupied by the act of
copulation, which I have heard put at two hours.

Eventually, however, the time arrives when the tusker proceeds
to action. First one fore foot, and then the other, is lifted off the
ground, and slowly swung from side to side across his front,
exactly as though he were deliberating from which foot to take
off. Then, levering himself up by his tusks, he places both fore
feet on the female's flanks, at the same time sinking down on his
hind legs to an almost sitting position. It is at this moment that
he shows the most violent sexual excitement. His organ, which
up till now has been bent back in the form of an arc, with its end
near the ground, and a few inches only in front of his hind feet,
is shot forward, and jerked up and down in a most violent manner.
It is flourished in the air, and often bent into the form of an S
like the thong of a whip, during which semen may or may not be
emitted. At one moment the organ appears outside the tusker's
shoulder, the next it is seen proti-uding out behind his hind legs.
It is kept in a state of continual agitation, and not for one single
moment is it still. Eventually it is jerked up straight forward
and hits the female organ from below. Penetration is then
immediately effected without further difficulty. The tusker then
raises himself until his hind legs are quite straight, his organ is
pushed home, and his fore feet are slid along the female's back
until they rest on her shoulders. Arrived at this position, he
begins to work very much after the fashion of a dog, and I
usually counted from six to eight horizontal motions. Connection
being then complete, the tusker lifts his tusks ofi" the female's
back and raises his head into the air. In this position he remains
for a few seconds, and then slowly withdraws his organ, letting
himself gently down to the ground in the same way that he
mounted, and qiiietly moves off. It is then that the female
shows her only signs of excitement : she trumpets softly as with
pleasure, thrusts forward her ears and stiffens her tail, her whole
conduct being indicative of pleasure and pride. The tusker is
throughout perfectly silent.

It is noteworthy that at no time does the tusker use his fore legs
to steady himself or to grip the female as does a dog or a stallion ;
the feet are invariably kept close together on the top of the female's
back, and I can quite believe the statement of tlie Burmans that
it makes no difference whether his feet are fettered or not.

I was unable to distinguish any sign by which I could tell when
a female was in season, but three out of the four females in camp
did come into season during the period of 10 days in February
during which they were under observation.

The operation was accurately timed on one occasion, and was
found to last exactly one minute.

From my own observations, I am convinced it is the female
that comes in season, and that until she does so come the tusker

P.Z.S.1903.VO1.1.P1.X1I

1/ 1 \.ri-»''

.<ry/

Z.D & S.J H. del.

Eale k Darnels son L*"'^ M.

CERATELLA. MINIMA.

1903.] ox CCELEXTERATA FROM ZANZIBAR. 113

will take no notice of her : also that he is rejuly whenever she is.
I can ofFei' no opinion as to whether the companionship of the
tusker tends to bi'ing the female into season or not.

I nuist draw attention to the way in which both the tusker and
female absolutely ignoi'ed oui' presence. As I have already st;ited,
the camera was set up within 10 yai-ds of the tethered female,
quite out in the open witliout any attempt at concealment ; and
though there were sometimes six or eight spectators walking about
and talking within this short distance, on no occasion did either
the tusker or the female take the slightest notice of us or even look
our way. Dii'cctly the act of copulation was over, the maliout
called out to the tusker, who, at the word of command, came towards
him, knelt down, and allowed himself to be mounted and ridden ofi".

As I have said, the female was always tethei'ed fore and aft
with a long chain, but this was solely with the object of prevent-
ing her swinging round and so getting out of focus ; but in no
single instance did she attempt to do this, nor was the chain once
needed except, of course, in the case of the first female who was
not in season.

7. On the Cffilenterata collected by Mr. C. Crosslaud in
Zanzibar. — I. Ceratella minima, n. sp. By Sydney J.
HiCKSON, M.A., F.R.S., F.Z.S., Beyer Professor of
Zoology in the Owens College, Manchester.
[Received December 16, 1902.]
(Plate XIII. ^)

Our knowledge of the remarkable family of Hydi-ozoa, the
Cei-atellada^, has been ably summarized in the memoir pul)lished
by Prof. Spencer in the Transactions of the Royal (Society of
Victoria, 1892 (4). The hithei'to recorded species are distributed
as follows: — Ceratella fasca : Coogee, Bondi (N.S.W.), Broughton
Island, Flinders Island, Lord Plowe Island. C. jiroca.mbeiis : Cape
of Good Hope, Natal. C. spinosa: Port Natal. Chltiiia ericoj)sis :
New Zealand. Dehitella atrorahens: Delagoa Bay. In brief,
the family has hitherto been known to occur only in Australasian
and S. Afi-ican waters.

The discovery of a new species of the genus Ceratella in the
ti'opical waters of the Zanziliar coast is in itself worthy of note,
but especially so in view of the fact that it is asso(;iated witli
corals, alcyonarians, and other animals charactei'istic of the
tropical belt of the East African coast. I am indebted to
Mr. C. Crosslaud, of Clare College, Cambi-idge, who collected the
three specimens during his recent exjiedition to Zanzibar, for
pel-mission to examine and describe them.

The African species of the family have not yet been accurately
described, but the accounts of Gray (1) and Carter (2) are

1 For fX]il!iimtioii of the I'late soc j). 11(3.

Proc. Zool. Soc— 1903, Vol. I. No. ^^111. 8

114 PROP. s. J. HicKSON ON [Feb. 3,

sufficient to show that the Zanzibar specimens should be described
as belonging to a new species.

They appear to be most closely related to the Australian
species Ceratella fusca, but diifer from it in one or two characters
which Spencer and his predecessors regarded as of generic
impoi'tance.

The genus was defined by Spencer as follows :— Colony
irregularly branching ; more or less expanded in one plane ;
growing from a creeping base. Main stem flattened, branches
i-ounded and beset with bracket-like hydi'ophores.

In C. ifninima from Zanzibar the main stem is not flattened
but perfectly cylindrical in form, and the hydrophores are so
extremely reduced or rudimentary that they are little moi-e than
ridges on the proximal lips of the hydi-opores. The branching,
moi-eover, appears to be strictly in one plane, and the terminal
branches are much more slender and delicate than in the other
species. Before passing on to the specific characters, the size of
the specimens must be considered.

The measm^ements are as follows : — ■

Specimen A. Specimen B. Specimen C.
mm. mm. mm.

Height of the colony 29 22 35

Maximum expanse of the

branches 65 38 50

Diameter of main stem 1"2 0*75 1

The colonies of Ceratellcc fihsca are from 11 to 5 inches in
height ; the largest specimens of C . procumbens described by Carter
were 1 1 inches long by 5 inches broad, and of C. spinosa 4~ inches
long by 2 broad. The height of Chitina ericopsis is 14 inches,
with a trunk (main stem) 1 inch in diameter. The size of
Dehitella is not given by Gray, but from the figures it may be
judged that it is larger than Ceratella fusca. From these figures
it is clear that the Zanzibar specimens are much smaller than the
average size of the adult colonies of the other species. Are they,
therefore, to be regarded as young colonies or as the representa-
tives of a dwarf species ? If they are young colonies, it is quite
possible that the main stem or ti-unk becomes somewhat compressed
in the plane of branching as the colony grows ; but the fact that
all the three specimens obtained are of appi'oximately the same
size, suggests that they have reached or nearly reached their
maximum growth. The dwarfing of the tropical species of a genus
that is pi'incipally distributed in temperate waters is not without
pai'allel in the group of Coelenterata. The very rudimentary
chai'acter of the hydrophores, however, cannot be explained by
the suggestion of immaturity, and must be regarded as of specific
importance. It is true that no gonophores have been discovered
in the specimens, but it is quite probable that, as in other
Coelenterates of the tropics, their production is rapid and strictly
seasonal, so that no ai-gument can be deduced fi'om this character,
either for oi- against the theory of juvenility.

The Colony. — The biunching is not very profuse, and strictly

1903.] CCELEXTERATA FROM ZANZIBAR. 115

confined to one plane (PI. XIII. fig. 1). The main stem and the
larger branches seem to have divided dichotomously with the pre-
dominance of the most favoiu'ed branch. The terminal branches
are veiy delicate, branches of 6 mm. in length gradually attenuating
from 0-2 mm. to 0"1 mm. in diameter. Each terminal branch ends in
a facultative gl•o^\•ing point, and it appears probable that the growth
is continuous. I have compai-ed my specimens from Zanzibar
with a beautifully preserved specimen of Ceratellafusca, for which
the Manchester Museum is indebted to Pi'of . Spencei- ; and I have
noticed that the Austi-alian species is much coarser in appearance,
especially in the region of the terminal branches. I have seen
nothing in the Zanzibar species corresponding to what Spencer
calls "The growing ends of the smaller branches" in Ceratella
fusca, which ai-e flattened in a plane at right angles to that in
which the general growth takes place and are entirely devoid of
zooids. It is possible that the diflei-ence may be accounted for
on the supposition that in Ceratella fasca the growth is seasonal
or periodic.

The branches of C. minima are invariably rounded. I have
seen no evidence of a compression or flattening in any region.
The sui'face is i-elatively smooth and fi-ee fi'om any spines. Tlio
hydrophores are represented by very narrow ridges on the
proximal border of the hydropores. The principal horny fibres,
I'unning longitudinally with a slightly spii'al twist, may be clearly
seen through the superficial ectoderm, and in the spirit-specimens
give a ribbed appearance to the surface (PI. XIII. fig. 2).

The Zooids are numerous on the terminal bi'anches, less
numei'ous on the thicker branches, and very scarce on the main
branches, as in Ceratella fusca. On the terminal branches they ai-e
ari'anged slightly to one side of the two lateral lines at intervals
of about one millimetre opposite or alternate to one another. A
few zooids occur more irregularly distributed.

On comparing such a terminal branch with one of Ceratella
fasca, it is at once apparent that in the Australian species the
zooids ai'e more numei'ous and much more irregulai'ly distributed
on all sides of the branch. Each fully expanded zooid projects
about 0'7 mm. from the hydropore, and is about 0*14 mm. in
diameter. It bears a variable number, but usually nine capitate
tentacles, each about O'l mm. iia length.

No gonophores were found on any of the three specimens I
have examined. The skeleton in the tei'minal branches consists
of one or two main longitudinal horny I'ods supporting numerous
looping and irregular bands, which maintain the cylindrical form
of theVanch (PI. XIII. fig. 3).

In the larger branches the main longitudinal rods are more
numerous, and, being chiefly superficial in position, give a slightly
spiiul, longitudinally striated or ril)l)e<l appearance to the surface,
in this respect ottering a markeil contrast to Ceratella fasca.

A series of sections through a small branch shows tliat there is
present a thin continuous coat of ectoderm covering the whole
branch, as described and figured bv Spencer in Ceralella fasca,

8*

116 ON CCELENTERATA FROM ZANZIBAR. [Feb. 3,

The ariungement of the canals and the general histology does not
appear to differ matei'ially from the desciiption given by the same
author ; but as the state of preservation of the Zanzibar specimens
was not perfect, a detailed and critical examination of the sections
was not made. I was fortunate enough to find, however, a few
nematocysts that were exploded but I'emained in situ. They are
(PL XIII. fig. 4) very similar in form to the small nematocysts
■of Millejjora, and exhibit a vesicle and neck O'Ol mm. in length,
;armed with four barbs or spines. The thread was invariably
brolcen, and I have no means of measuring its extreme length.
The endoderm of the tentacles is solid.

The diagnosis of the species is as follows : —

'Ceratella minima, sp. nov.

Colony probably erect, branching strictly in one plane, iri'egulaily
and not very profusely. The main stems and all the branches
cylindrical in form. Hydrophores very slightly developed.
Skeleton consisting of dark brown longitudinally disposed horny
fibres, united by loops and bands, forming in the larger branches
a dense and firm but flexible skeletal plexus.

Zooids situated slightly to one side of the plane of branching of
the colony, alternately or in pairs, at distances of about 1 mm. on
the terminal branches. Lai-gest colony 29 mm. in height, with a
maximum expanse of 65 mm.

Locality. Zanzibar, shallow water.

Literature.

1. J. E. Gray. — Notes on the Ceratelladte, a Family of Keratose

Sponges. Proc. Zool. Soc. 1868, p. 575.

2. H. J. Carter. — Transformation of an entire Shell into

Chitinous Structure by the Polype Ilydractinia, with short
descriptions of the Polypidoms of five other species. Ann.
& Mag. Nat. Hist. ser. 4, 1873, xi. p. 1.

3. W. M. Bale. — Some new and rare Hydroida in the Australian

Museum. Proc. Linn. Soc. N. S.Wales, 1889, p. 748.

4. W. B. Spencer. — On the Structure of Gei^atella fusca. Trans.

Royal Soc. of Victoria, 1892, p. 8.

EXPLANATION OF PLATE XIII.

Pig. 1. Drawing of the whole colony of Ceratella minima (specimen A) twice the
natural size. The base of attachment is covered by an encrusting polyzoou.

Pig. 2. A portion of a large branch and smaller branchiet taken from the region
marked * in fig. 1, more highly magnified, showing the polyps P partiallj'-
extended, bearing an irregular number of knobbed tentacles, t.t. At 7^./^. are
seen the rudimentary hj'drophores. The figures show the general arrangement
of the fibres of the horny skeleton. In the large branch they are closely
crowded togetlier. In the smaller branches the principal fibres are separated
by considerable spaces but maintain a parallel arrangement.

Pig. 3. A still smaller branch more highly magnified, in which there is only one
primary longitudinal fibre.

Pig. 4. One of the nematocysts of the tentacles. The diameter of the cyst is about
O'Ol mm.

1903.] ox THE PLAXKTOX OF THE FAEROE CIIAXXEL. 117

8. Contriljutions to our Knowledge o£ the Plankton of the
Faeroe C-hannel. — No. VIII. ^ By G. Heubert Fowlkii,
B.A., Ph.D., F.Z.S.

[Koceivod Dcccmlicr 20, 1902.]

(Text-figures 13-17.)

The present pai:)er contains notes {in some cases due to tlio
vnlued help of friends) on Beroe, Arachnactis, Podon, the Ostra<:'oda.,
the Oopepoda, the Anipliipoda, and the Sehizopoda., captured by
H.M.S. ' Research' in 1896 and 1897 in the Faeroe Channel.

Otenopiiora.
Beroe cucumis Fabricius.

This species, cliai-actei'istic of cold Arctic curi-ents ', was taken in
the following hauls : —

16«ii., 300 to 170 fathoms, seven specimens.

IS e, 400 to ? fathoms, one fragment.

20 f^, 500 to 400 fathoms, one specimen.

Most specimens showed the characteristic brick-red or rose tint,
and though much battered and in some cases inverted, were
refei'abl'e with a fair amount of certainty to this species.

A N T II o z A,
ARACnxACTis ALBiDA M. Sars.

Some information as to the develojmiental succession of the
mesenteries in this form was given in ISTo. III. of this series.
Since its publication, I am glad to say that it has been substanti-
ally corroborated by Prof, van Beneden ''.

He agrees with my suggestion to sepa.rate the Channel and
ISTOrth Sea Arachnactis from alh'ida of the Faeroe Channel, and
describes it under the name of lloydii, under the idea, Avhich is
probable, but at present improved, that it will be shown eventu-
ally to be the larva of Cereanthus lloydii. Till this has been
proved, I venture to think it better to retain my pro\isional naino
of hournei for this form.

The occurrences of A. albida are shown in the table (p. 118) :
it occurred in over 61 per cent, of epiplnnkton hauls, never in a
mesoplnnktou haul, and may fairly be taken to be a purely
epiplankton form. It was present in considerable quantity, as
many as 50 specimens having been taken in one Jiatd.

1 The references to previous papers in (lie Society's Proccedinirs are : — No. T., 1800
p. 991 ; No. II.. 1897, p. o2.3 : No. III., 1S97. p. m:S ; No. IV.,1K98, p.olO; IS',,. V.,
1898, p. 550; No. VI., ISitS, p. 5C.7 ; Xo. VII., 1898, ]>. lOlC. I regret that various
circumstances, mostly beyond my control, have caused so irreat a lapse ot time
between this paper and No. VII.

- Chun : ' Die Ctenoi)horen der Plank ton-Kxjiedition,' p. 2li.
3 Van Ueneden : ' Anthozoaircs de la Plankton-Expedition.'

118

DR. G. HERBERT FOWLER ON THE

[Feb. 3,

■^

=0

o

1

8

1

■i

« r-"

1

-2 c

'«

a

a 3

^

a

CO

h V-

S g

a

^1
11

o

S

5*

Jog

r

11
3^

11a.

100-+0

+

116.

■(-

+

lie.

30-0

■f

+

12 S.

10-0

+

+

12 c.

•)- +

+

12 d.

130-±0

..*

+

12/.

PlO-0

4-

13 e.

2-0

+ +

13/.

+ +

13 A.

fc

13 i.

100-0

+ +

+

o

H

13yfc.

2-0

+ +

M

13 Z.

14.

+

Hi

15 a.

2-0

...

Pi

15 6.

+

+

W

15 c^.

16 6.
16 e.
17.

18 a.

19 c.

19 d.

20 e.
20/
20 y.

4-0

3-0
4-0
10-0

40-0

+

+ +

+

+

+
+

+
+
+

+
+

+

12 a.

±350-±150

12 e.

450-320

+

>q

13 d:.

400-270

+

O

135'.

465-335

+

+

M

16 a i.

350-220

+

+

»
^

16aii.

300-170

+

+

Hi
o

18 6.

530-400

+

19 a.

480-350

+

+

CQ

20 a.

200-100

+

+

^

20 6.
20 c.

300-200
400-300

+

+

+

20 i.

500-400

+

+

+

+

^•

13 a 6.

300-O

+

+

+

t+-<

13 e.

400-?

+

+

+

+

■^

15 c.

530-0

+

+

o

p

19 6.

480-0

+

+

1903.]

PLAXKTOX OF THE FAEROE CHAXXEL,

119

822

:^

^s

r

8^

e"

B

8

s

2

1

S

« s5
S 3

J*

'§>

5^'s

S

8

S

IS «

^^

id

§^i

S3

1"^

Si Si

1-'^

4''«

=^ci

,^,s

^-o

Es

E-i

§

fei

&H

+

+

+

+

+

+

+
+

+

+

+

+

+

+

+
+

+

+

+

+
+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+
+

+

+
+

+
+

+

+

+

+

+

+

+

+

.+

+

+

+

+

+

+
+

+

+

+

+

+

120

DR. G. HERBERT FOWLER ON THE

[Feb. 3,

Phyllopoda.

PODON INTERMEDIUS Lilljeborg.

My friend the Rev. T. R. R. Stebbing, F.R.S., was kind enongh
to identify and count the specimens of this species. It would no
doubt have occurred in more of the svirface hauls had not my
finest tow-net been devoted to the collection of Diatoms for the
Scottish Fisheiy Board. It is no doubt a purely epiplanktonic
form in the Faeroe Channel ; a single sjoecimen only was taken at
20 d (500-400 fathoms) as against about 106 specimens in 7 hauls
at the svirface ; the single deep specimen was probably a sinking
corpse.

This species is, I believe, known only from the surface ^ ; it
ranges over the Baltic, Norwegian coast up to Vadso, Denmark,
Boulogne, Concarneau, Trieste.

OSTRACODA.

The representatives of this order belonged exclusively to the
Halocypridse, and were mostly taken in the mesoplankton. Only
one sjDecies, Conchoecia maxima, occui'red sufficiently often to allow
of a generalisation as to its horizon.

The identification of a Halocypiid is rarely satisfactory without
dissection of the mouth-parts, which means destruction of the
specimen. I have, however, dissected a considerable number, and
feel at all doubtful only in the case of G. porrecta ; I have pre-
ferred, however, to leave the specimens under this species leather
than create a new species on the strength of slight differences
in the armature of the mandible.

The table following shows those svirface and deep-water hauls,
made by Professor Chun between Finistere and the Canary Islands
with open vertical nets^, which contained the same species as

g"

e

Haul.

Horizon in fathoms.

.si

1 =

8

li

^ 2
5'^

O

o ■

Ph

fel

6

II.

546 to

*

rJ2

III.

819 to

*

*

-§

#

IV.

546 to

*

*

*

*

VII.

873 to
Surface.

*

*

>
. *

1 J. de Guerne: Bull. Soc. Zool. Trance, xii. 341 (1887). W. Lilljeborg:
' Cladocera Sueciaj,' Upsala, 1900, 4to.

2 Chun : SB. kon. preuss. Akad. Wissensch. (1889 xxx. Claus : ' Die Halocy.
priden,' Wicn, 1891, 4to.

1903.] PLAXKTON OF TUE FAEKOE CUANXEU 121

were captured by the ' Research ' in the Faeroe Channel ; they
are cited in the text by Roman numerals.

CONCHCECIA MAXIMA Brady (fe Norman,

As I have previously pointed out \ tliis species appears to be
purely mesoplanktonic, in the latitude at any rate of the Faeroe
Channel. It occurred in 50 per cent, of the mesophvnkton liauls,
and in three hauls which began at or over 300 fathoms and finished
at the surface ; it was not captured once in hauls between 100
fathoms and the sui-face {cf. table, p. 118). The species was
fairly common.

The record of previous captures Avas cited in the second paper of
this series \ and also indicated a mesoplanktonic habit in subarctic
regions, but it ls not surprising that in yet colder waters it should
appear at the surface. At 84° 32' N., 76° E., it was captured
with a surface-net by the ' Fram'",and is recorded as abundant in
most of the samples of Crustacea from this voyage ''.

CONCHOECIA HYALOPHYLLUM ClaUS.

Twelve specimens in haul 13 i, 100 fathoms to surface. Five
specimens too small for satisfactory identification, but perhaps
referable to this species, occurred in hauls 20 c and 20 d.

Glaus ^ records as other occui-i'ences Chun's haul IV., Ischia at
492 fathoms, Oi'otava at the sui'face.

? COXCHCECIA PORRECTA ClauS.

Numerous specimens from 20 a bod and one from 13 i. Claus
(oj). cit.) records it from Chun's hauls II., III., IV., VII.

CoxcHCECiA BOREALis G. O. Sars.

A single specimen in haul 19 «, 480 to 350 fathoms. Recorded
previously from 250 to 300 fathoms at the Lofoten Islands ', and
from Trondhjem Fjord at 150 fathoms®. This appears to be a
purely cold-water form.

Paracoxch(ecia oblonga Claus.

Six specimens in haul 20 c, four in 20 (L Claus {op. cit. p. 64)
cites this species as from Chun's hauls III. and IV. ; and states
that it also occurs at the surface, but without giving authoiity or
details. He remarks on the probable identity of this species with
G. W. Midler's variabilis', a suggestion with which Miiller seems
to agree "*. This would extend the distrilmtion considerably, as
Miiller '' records it, from the ' Vettor Pisani ' collections of

1 Proc. Zool. Soc. 1897, p. 523.

2 F. Nansen : Norwegian North Polar E.vpuditioii. G. 0. Sai-s: Cnistacua, p. 11.

3 Id. ibid. p. 137.

■1 C. Claus, op. cit. p. 61.

5 G. O. Sars : Fori). Vid.-Selsk. Christiania (1865), vol. 1866, p. 120.

6 G. S. Brady & A. M. Nonnan : Trans. Koy. Dublin Soc. (2) v. p. 086.
' G. W. Miiller: " Ueber Halocypriden," Zool. Jahrb. Syst. v. p. 273.

8 G. W. Miiller : 'Ostracoden des Golfes vou Neapel,' p. 229.

122 DR. G. HERBERT FOWLER ON THE [Feb. 3,

Chierchia, as occurring at various points in the tropics at depths
between 382 and 546 fathoms ; it occurs also in the Gulf of
Naples. Brady ' records it as having been taken by the ' Chal-
lenger ' Expedition off Kandavu, Fiji, and between Marion and
Crozet Islands at unrecorded depths.

Halocypria globosa Claus.

Six specimens in haul 1 3 i.

This species is known from the surface and at various depths
in the Atlantic ^, and is recorded from Gibraltar as taken by the
' Yettor Pisani ' ^ Of specimens taken by the ' Challenger,' the
record was in one instance lost * ; other specimens were captured
at the surface between Api and Cape York \ It seems to be a
form widely distiibuted both vertically and horizontally.

CONCHCECILLA DAPHNOIDES ClauS,

Only three complete specimens, and one empty carapace, refer-
able to this genus were obtained. The specimens on which Claus
founded the genus (with this single species) were all young males :
larger specimens of the genus, including females, were obtained by
Sir John Murray on H.M.S. 'Triton' in 1882, from the Cold
Area of the Faeroe Channel, and were described by Canon
Norman and Dr. Brady under the specific name of lacerta, not
without the " suspicion that they may perhaps belong to the adult
form of C. daphnoides " ^. My own specimens were too few to
settle the point ; but as the two smaller specimens most resembled
in outline the figvire of Claus, and the two largest that of Brady
and Norman, I have left them provisionally under the older specific
name.

In addition to Chun's hauls III, and lY., it has been captured
at 200 fathoms off Achill Head {daphnoides) ^ the Faeroe Channel
as above {lacerta), and off Kandavu, Fiji, at an unrecorded depth ^
{dap)hnoides).

COPEPODA.

Mr. I. C. Thompson was kind enough to report on the
Copepoda in No. lY. of this series of papers '^. Since that date,
the arrangement of three then doubtful hauls has required modi-
fication : 12 a, which was suspected at the time of capture to have
remained open too long, proves to have no apparent contamination
of undoubted surface forms, and has been moved to the Meso-
plankton ; 1 3 e was also suspected, in this case with justice, as

1 G. S. Brad}' : " Myodocopa of the ' Challenger ' Expedition," Trans. Zool. Soc.
xiv. p. 95.

2 C. Claus, op. cit. p. 79.

3 G. W. Miiller : Zool. Jahrb. Sysfc. v. p. 270.

* G. S. Brady & A. M. Norman : Trans. Roy. Dublin Soc. (2) v. p. 705.

5 G. S. Brady : Trans. Zool. Soc. xiv. p. 97.

6 G. S. Brady & A. M. Norman : Trans. Roy. Dublin Soc. (2) v. p. 697.

7 G. S. Brady : Trans. Zool. Soc. xiv. p. 95.

8 Proc. Zool. Soc. 1898, p. 540.

190.'3.] PLAXKTON OF THE FAEROE C'lIAXXEL. 123

containing several undoubtedly epiplanktonic species (e. g. Arach-
nactis albida), and has been relegated to the " doubtful" category,
closure of the net not having taken place at the proper time ;
12/ was known to be epiplanktonic, although there was some
doubt as to the exact depth at which it had been towed. I have
therefore reprinted in the table (pp. 118, 119) the captures of the
seven fonns which were taken at least six times, sufficiently often
to give ajDproximate data for an estimate of their vertical distri-
bution. In discussing this question, I gave a short table on
p. 546 showing the occurrences of these seven species expressed
in percentages of those Epiplankton and Mesoplankton hauls
which contained Gopepoda ; this can now be amendetl as follows,
omitting the four doubtful hauls from the calculation : — -

Total hauls containing Copepoda.

!

Epiplankton.

M(

\
jsoplankton.

Calamisfinmarchicus occurr

ed in

83 "/n

and

in

91 "/o

^ Eucalanns attennatas

22 „

41 „

lEiicJueta norvegica

11 »

75 „

3[ctridia longa

22

75 „

'^ Pleuromma ahdomiiiale

"5 ,','

58 „

Acartia clausii

33 „

25 „

Temora longicornis

33 „

„

The amended table is in harmony with the conclusions dnxwn
from the former, as to the vertical distribution of these forms,
except in the case of Acartia clausii, the question of which was
expressly resei-ved {pp. cif. p. 549).

Since the publication of Mr. Thompson's report, I found that
Dr. R. Norris Wolfenden was making an exhaustive study of the
fauna of the Faeroe Channel, and natumlly j^laced my collection
at his disposal. He has been kind enough to f ui^nish the following
notes on new and other species, with some of which he has
already dealt briefly elsewhere ^, Exact data of depth &c. were
not always available, as by the time that Dr. Wolfenden received
the specimens all the epiplankton hauls of a station had in many
cases been put together in one bottle, all the mesoplankton hauls
in another, for economy of space.

Pleuromma robustum Dahl, Zool. Anzeig. v. p. 16 (1893).

"This is the common Fleioromma of the Faeroe Channel, and
was found to be present in considerable numbers in Dr. Fowler's
collection, PL ahdominale occurring much more rarely. In these
northern latitudes it nlmost entirely replaces the latter species.
It is readily distingviishable by the horseshoe-shaped mass of red
pigment which is present in the anterior and inferior portion of
the head at the base of the mouth-organs, and in the male by

^ Dr. Wolfenden points out that as, according to his wide experience, T^ucalanus
attenuatus is nowhere met with in the Faeroe Channel, unless perhaps quite
exceptionally, these figures probably refer to Eucalantis elonijatiis. Similarly, for
the species Pleuromma ahdominale should probably be substituted P/. robustum.

- R. N. Wolfenden : Journ. Marine Biol. Assoc. (1902), vi. p. 344.

124 DR. G. HERBERT FOWLER ON THE [Feb. 3,

the clasping antenna being on the left side, and the pigment-
spot invariably on the right side. The secend feet in both
male and female have the characteristic notch and hook on both
limbs. Length 3-4 mm. Haul 20 a (200 to 100 fms.) and several
others."

HetergcHtEta zetesios c5" , Wolfenden, op. cit. p. 367.

" The head is like E. jyapilUgera Gbt. Though the end joints
of both anterior antennse were broken off, the 19 joints left had
a length of over 4 mm., with the geniculation between the 18th
and 19th segments. The anterior antennae were therefore much
longer than the whole animal, which was 3*5 mm. There was
considerable asymmetry of the furcal segments, that on the left
being much the longest and broadest. The anterior foot-jaw had
one thick hooked bristle on the 5th lobe, but no " tooth-comb "
bristle, and the 5th feet were peculiar and unlike those of any
other Heterochceta, displaying an upright and stiff process of the
2nd basal joint, armed with fine stiff hairs on the inner aspect
(like a " tooth-comb "), and the proximal inner margin of the 1st
joint of the exopodite with a protuberance armed with 4 teeth.
The 2nd basal joint of the foot of the opposite side is armed distally
with short stiff bristles. It could be only the male of H. grimaldii
or of H. longicornis, neither of which is yet known, or of H. major
(Dahl). The latter and R. grimaldii are very large (5-10 mm.),
and though II. zetesios resembles II. longicornis in some points, it
is perhaps better for the present to distinguish it as a new species.
Only one example was met with in Dr. Fowler's collection, in
haul 20 « (200 to 100 fathoms)."

^GiSTHUS ATLANTicus Wolfenden, op. cit. p. 364.

" The occurrence of an example of this genus in the Faeroe
Channel is remarkable. This specimen was found in the collection
made by Dr. Fowler as 20 a (200 to 100 fathoms). It had a
total length of 1*45 mm., a 6-segmented anterior antenna with
very long and peculiar sensory processes. It has distinct differ-
ences from uEg. 'mucronatus or ^Eg. acideatv.s Gbt., and also from
the species described by Scott from the Gulf of Guinea as
jEgisthus longirostris."

LucicuTiA MAGNA Wolfenden, sp. n. c? •

"A single specimen found in Dr. Fowler's collection from 19 a
(480 to 350 fathoms), of 3*54 mm. length, was apparently new.
The anterior antennas were larger than the whole body, by the
terminal one and a half joints. The endopodite of the 1st foot
was two- jointed. The right 5th foot has a strong spiny process
on the inner side of the 2nd basal and the exopodite of two seg-
ments ; the endopodite and exopodite of the left 5th foot being
■ each of three segments. The size alone distinguishes it from the
males of any other known species, only L. grandis being larger."

1903.] PLAXKTOX OF THE FAEROE CHAXKEL. 125

AuGAPTiLus ZETESios Wolfeuden, op. cit. p. 369,

"One specimen only was found, in the bottle marked 19 a. c.
Another sijecimen occurred in the sample marked 20."

EucALANus CRASsus Gbt. 1888, Atti Ace. Line., and 1892, Fauna
u. Flora ISTeapel, v. p. 1 9.

" This species was of not infrequent occurrence, especially in
the bottles marked B 1 and 13 A- (2 to 0) and 13 i (100 to 0).

" The writer also has frequently noted its occurrence in the
Faeroe Channel."

Gaetaxus major Wolfenden, sp. n.

" Two examples of this genus {Gaetanus) were found in
Dr. Fowler's collection marked 19 « (480 to 350 fathoms). The
copepod very greatly resembled Gaetaims armiger Gbt., but the
anterior autennc-e were longer, i-eaching beyond the f urea by the
length of the last joint ; the spines of the last thoracic segment
were comparatively shorter, the 1st abdominal segment and tlie
anal segment shorter, and the furcal segments only as Ion"- as
broad (longer than broad in G. armiger), and each abdominal
segment had a row of pectinations on the posterior border. The
saws of the swimming-feet possessed more teeth ; the abdomen
Avas not nearly half the length of the cephalothorax, and the
whole length of the animal was 5-3 mm. In all these points it
difiered from the typical G. armiger, the size of which reaches
only about 3 mm., and justifies its being made into a sejjarate
species."

Gaidius Gbt. 1895, Bull. Mus. Harvard.

" A good many examples of this genus occurred in the deep-
water collections of Dr. Fowler, e. g. in the bottle marked Meso-
plankton 20 (500 to 100 fathoms).

_ There is no doubt that in the Faeroe Channel there are two
kinds of Gaidius — one agreeing in every particular with the
Gaidias 2iangens oi Giesbrecht; the other, a larger s^oecies, also
difiering in the segmentation of the 1st and 2nd feet. Gaidius
2)iingeas Gbt. has the exopodite of the 1st foot with only two
segments ;ind the endopodite of the 2nd foot with only one seo--
ment; whereas the northern species has a 3-jointed exopodite
of the 1st foot and a two-jointed endopodite of the 2nd foot.
There are other minor differences. In size the northern species is
much larger ; Gaidias jningens Gbt. being 4| mm,, tis compared
with about 3 mm. The Chiridius temcis2nnis~'oi G. 0. Sars is the
same species, all being characterized by the peculiar series of
lamellar ajipendages of the basipodite of the 4th foot."

PsEUDAETiDEUS ARMATUS "Wolfenden.

" Some examples of this species occurred in Dr. Fowler's col-
lection. It was drawn and described (in IM8S. only) before the

126 DR. G. HERBERT FOWLER ON THE [Feb. 3,

writer became acquainted with the recent work of Prof. G. 0. Sars,
who figured and described the species as Chiridius armatus. The
writer has pubhshed reasons why this generic name should not
be used (see Rep. of the Brit. Assoc. 1892, " A proposed Revision
of the Subfamily Aetidiince "), as it is not a Chiridius."

EuCHiRELLA CARINATA Wolfenden, op. cit. p. 366.

" One example of this species, measuring 3*54 mm. in length,
was found in Dr. Fowler's collection in the bottle marked 20.

"These new species will be described in full, along with the
drawings, in the writer's monograph which is in hand."

Amphipoda.

I am indebted to the Rev. T. R. R. Stabbing for help in the
determination of some of these forms. Only one species occurred
in sufhcient quantity and with sufficient frequency to enable
deductions as to its habitat in the Faeroe Channel being drawn,
namely

Parathemisto oblivia 'Krojer = abi/ssorum Boeck.

I have already discussed the distribution of this form at some
length ^ ; to the records there given must be added two stations
of the ' National ' expedition — N. of the Hebrides, and S.W. of
Iceland '^ — and 1 2 stations along the route of the ' Fram ' ^ Canon
Norman ajso cites Bonnier as having taken it at 950 meti-es in
the Bay of Biscay *.

It is apparent from the table (p. 119) that the species is a
true member of the Mesoplankton in this locality, having been
captured in 66 per cent, of the deep hauls ; it rises to the surface
at midnight, the only occasion out of 26 Epiplankton hauls being
at that hour (haul 1 5 d).

Cyclocaris guilelmi Chevreux.

Two specimens of a Cyclocaris were obtained in haul 20 d,
between 500 and 400 fathoms. Mr. Stebbing informs me that
they agi'ee undoubtedly with the above species, captured in a net
sunk to 600 fathoms near the Lofoten Islands by the ' Princesse
Alice ' '% the largest of the six specimens being about 1 2 mm. in
length. My largest specimen, a female, would have measured
about 20 mm. if straight.

Another species of this genus, Cyclocaris tahitensis Stebbing,
was described ^ from a single specimen taken by the ' Challenger '
off Tahiti, apparently at the suiface ''.

1 Proc. Zool. Soc. 1898, pp. 583-585.

2 J. Vosseler : Amphipodeii der Plankton-Expedition, p. 80.

^ P. Nansen : Nonvegian Nortli Polar Expedition. Crustacea, b}' G. 0. Sars,
p. 14.

4 A. M. Norman : Ann. Mag. N. H. (7) v. p. 131.

5 E. Clievreux : Bull. Soc. Zool. France, xxiv. (1899), p. 148.

* T. R. R. Stebbing : Chall. Rep. Zool., Amphipoda, p. 661, pi. xvii.
^ J. Murray : Chall. Rep., Summary of Results, p. 1077.

^^^'^•] PLAXKTOX OF THE PAEROE CHANNEL. 127

Mr Stel)biug points out "that tliis species has been a-ain
figured in great detail hy Pi-ofe,ssor Sars \ and tluit Dr. Nornmn ■=

T \T^ if , !°'* f'^"'"' ^''''''''' I't^I'i'^s^^ted his Ci/clocarisfaroensis.
in Mr. btebbings opinion the diflerences between the two sets of
hgures are purely casual, depending ou individual or accidental
conditions of the specimens examined. Further, wliile acceptino-
provisionally the distinction between the boreal form and the
Cliallenger C. tahitensis, he agrees with the view thus exin-essed
by i)r. Norman, 'so remarkable is the resemblance, that the
diflerences seem scarcely varietal ; but I hesitate to unite a form
Sliti"" Channel with one from so distant a locality as

Canon Norman does not appear to have known of M. Chevreux's
name (which apparently has priority); his specimens were captured
tTa i'!i '^^'V. "' *^'® ^'^^^'^^ Channel in 1882, at a dei.th of
640 fathoms Professor Sars's specimens were apparently from
various depths along the course of the ' Fram,' the smallest
specimens at the le-A&t depths.

This species (omitting tahitensis as specifically distinct) is
evidently a Polar and deep-water form.

TllYPHAXA MALMI Boeck.

A single specimen from haul 13 i, 100 fathoms to the surface,
was Identified by Mr. Stebbing. The previous records of its
occurrence are: Hardangerfjord, 100 fathoms'; from Ninsen-
jord to Hardangerfjord^; Folgero, Sunde, Foldenfjord, 80 to
100 fathoms fears ' regards T. nordeuskioldi Boeck and T. boecki
fetebbmg as being the males of T. malmi. Of these the former *
was described from the Sofia Experlition as oflHhe Faeroe Islands
at 65 N., the latter ^ from 18° 8' N., 30° 5' W., at the surface.

EuTHEMiSTO sp. indet.

A number of specimens too much broken for recognition
occurred m the haul 1 5 c^ at midnight. eco^nition

EuTHEMiSTo BisnxosA Boeck.

Occurred in 15 c (530 to fathoms) and Vo d (surface, at mid-
night). The records of this species were exclusively Arctic, till

tJZ '^'^9 «T .1^ the 'National' expedition in the Sargasso Sea
between 218 fathoms and the surface.'

EUTHEMISTO COMPRESSA Goes.

A single specimen at 12 6 (450 to 320 fathoms). This also was
believed to be exclusively an Arctic form, but has now been

pis. H: & "f" '- ^'''''^'-^'' ^°^'l' l'«^-' Expedition. G. O. 8ar. : C.-ustacca, p. 20,

3 i- ^k ^"vr""/- "^""i ^'*- ^- "• (' ) '-■ V- 197, 1.1. vi.

, "°«,f -■ J'kaiidiiiaviske og Arktiske Ampliipoda,' 1872, p 92

;« C. Hovalhus: Vega-Exped. Veteiisk. laktta- iv. p 573 '

•' hurs: Cnistafi'M of Norway: Anii.liipo.la, i;p. 17 (189U)

" T. K. II. Stebbing: Chull. Itop. AiuphipoJa, p. 1539

128 DR. G. HERBERT FOWLER ON THE [Feb, 3,

I'ecoi'ded from the Antarctic region, as well as from the Gulf-
Stream pi-oper, the Sargasso >Sea, and South Equatorial Drift ^.

EUTHEMISTO LIBELLULA Mandt.

A single specimen in haul 15 cZ at the surface at midnight. It
is widely distributed over the Arctic Seas, but, unlike the two
foregoing species, was not taken in southern waters by the
' National.'

SCINA BOREALIS G. 0. SarS.

A single specimen from 13 g (465 to 335 fathoms) was acci-
dentally included among the Gopepoda sent to Mi\ Thompson
and identified by Mr. A. O.Walker. According to Ganon Norman ",
the disti'ibution of this species is mesoplanktonic, mostly northern,
but also in the Bay of Biscay (960 metres). It reaches to about
80° N.'

SCHIZOPODA.

The Schizopoda captured belong exclusively to the Euphausiacea,
and are referable to only three species. Several forms which
might have been reasonably expected among the captures were
absent. A list of the British species with their distribution is
given by Ganon Noiinan, in his paper on Bi-itish Lophogastridse
and Euphausiidse *.

Thysanoessa loxgicaudata Kroyer.

A considerable number of specimens of this species were
captured : it appeared to be the commonest Schizopod of the
Faeroe Ghannel at the time. The synonymy appears to be
Tliysanoi^oda longicaiidata Kroyer = Thysanoessa tenera Sars =
Thysanoessa longicaudata of Hansen, Norman, Ortmann, &c.

The species ranges from the West Goast of Norway right across
to Greenland and into the Labi-ador current ; the ' National ' " ceased
to take it (after almost daily captures up to that moment) from
the date of entering the warm water of the Gulf-Stream (' Florida-
strom '). Sars **, who described the species fi'om deep water in
the Varanger Fjord, records it also as from the surface at four
stations between Norway and Jan Mayen on the cruise of the
' Voringen ' '^. It appears, therefore, to be essentially a cold-
water species, an Arctic type-form, and was captured by the
' Fram ' \

It has, however, been recorded twice from British coasts °.

^ Vosseler : Amphipoden der Plankton-Expedition, p. 86.

2 A. M. Norman : Ann. Mag. N. H. (7) v. p. 135.

^ F. Nansen, op. cit. p. 19.

■* A. M. Norman : Ann. Mag. Nat. Hist. (6) ix. p. 454.

5 A. Ortmann : ' Decapodcn und Scliizopoden der Plankton-Expedition,' p. 14.

6 G. 0. Sars: Porliandlinger Videnskabs-Selskahet (Christiania), 1882, no. 18, p. 53.
' G. 0. Sars : Norwegian North Atlantic Expedition (Crustacea), pt. ii. p. 13.

^ ¥. Nansen : Norwegian North Polar Expedition. G. O. Sars : Crustacea, p. 14.
" A. M. Norman : Ann. Mag. Nat. Hist. (6) ix. p. 463 ; and the papers there cited.

1903.] PLANKTON OF THE FAEROE CHANNEL. 120

Once it occuiTcd in cnoi'inous quantity in St. Andrew's JJay in
company with Nyctiphanes norvegica, once at Rodcriii- witl> a
similar swarm of E athemisto comj)ressa, the lattei- l)eing also an
Arctic type-form '. In both these cases it is probable that the
creatures had been driven down the North Sea by a .strong southerly
current, in the manner which I have already suggested '" foi*
Parathemisto ohlivia; and it has thei-efore no more i-ight to lie
i-egai-ded as a " British " species than an occasional Velella ov
lanthina bi'ought up by the North Atlantic Drift to our shoi'es.
According to Ortmann {pp. cit.), the ' National ' haids gave no
indication of the vertical distribution of this species.

It will appear from the ta))lo (p. 119) that the ' Research ' was
more foi'tunate, and the haids point to its having a distinct
pi'efei'ence for the mesoplankton in the Faeroe Channel. Like
(at any lute some) other mesoplaidvtonic species, it rises to the
surface at night. Specimens with adidt chaiucters were captiired
in 19 per cent, of epiplankton hauls, but in 06 per cent, of meso-
plankton haxds. On the other hand, lai'val and j'ost-lai'val stages,
apparently refeiulile to this species, wei'C obtained in 38 per cent.
of epiplankton hauls, but oidy in one mesoplankton haul, and
that one terminating near the 100 fathoms. The species, there-
fore, appears to be epiplanktonic when young, mesoplankton ic
when adult, so far as these observations go and in the Faeroe
Channel at this time of year. In seeking deepei" (coldei-) water
in this locality, it follows what appeal's to be the practice of other
Arctic type-forms when they meet the warmer water of the North
Atlantic J)i-ift. That this w^as not appai-ent from the results of
the ' National ' is probably due to the fact that from the Hebi'ides
almost up to the moment of coming into the Gulf-Stream the
vessel w'as in far colder surface-water than that of the Faeroe
Channel in summer.

The larva2 mentioned abo\"e ranged fiom an early Cahjptopis
stage up to the adult condition. It was not, of course, possible to
derive them all with cei'tainty from Thysaiioessa longicaudata ; but
the majority may l)e safely I'efei'red to this species, not only
because the adults captured were far in excess of any other
Euphausid, but also because the larva; could be traced gradually
thi'ough successive stages back to the Cahjptopis. The meta-
morphoses of this species follow the lines indicated by Sars ^ for
Nyctiphanes, Eujyhausia, and Thysanopoda.

As Paul Mayer' has shown, the spination of the telson of
Malacostracan la.rva; yields a character important both for ph5do-
geny and for diagnosis. It has not as yet, I think, been pointed
out that the condition of the telson in Euphausiida? affords a
further argument for the view maintained by Pxias' and othei'S,

' C. Chun : ' ]?czichui)gcn '/.wisclicn dcni avktisclicn nml aiit;ukfisilirn I'lnnktciii."
Stuttgart, 1897, 8vo, p. 30.
- Proc. Zool. Soc. 1898, p. 583.
3 G. 0. Sars : Cliall. K.'p. Zool. xiii. (Schizopoda).
•• P. Mayer: Jpnaischc Zcitsclirift, xi. (1877), p. 210 ct seqq.
^ .J. E. V. Boas: Morpliologisches Jahrbucli, viii. p. tSo.

Proc. Zool. Soc— 1903, Vol. I. No. IX. 9

130 DR. G, HERBERT FOWLER ON THE [Feb. 3,

that this family takes its origin very near to the root of the
Decapodan stem, and that it has far closer affinities with the
latter than with the Mysidse. In Nyctiphanes ^ and Euphausia ^,
and possibly in other genera also, the youngest Calyptopis-laxvse
show seven spines on each side of the telson ; unlike all other
Schizopoda, so far as is known, except perhaps Lophogaster, they
are thus in absolute accord with the " urspriingliche Borstenzahl
7 + 7 " which Mayer attributes to the primitive Macruran and
Brachyuran. In Euphausia, Nyctiphanes, Thysanopoda, and
JVematoscelis according to Sars (op. cit.), and in Thysanoessa,
the number is increased at later stages by a median terminal
spine, which, like the others, is jointed to the telson. Accepting
Mayer's enumeration of the spines from the middle line outwards,
and styling the median azygos spine of the Euphausiidse as 0, —
spines 7 are found in the adult Thysanoessa about one-third of
the length of the telson from the root ; spines 6 at about two-
thirds of its length from the root ; spines 5 are lost ; spines 4
persist as the large lateral jointed spines near the end of the adult
telson^; and spines 3, 2, 1, disappear altogether in the course
of development. On page 131, I have illustrated four stages in
this reduction omitted by Sars, of which fig. 15 does not quite
bear out his description : these show the disappearance of the
median spine 0, and the commencement of a new unjointed
growth of the telson backwards, to form the lanceolate tip of the
adult. The character of the telson and the presence of this
median spine will apparently form a good criterion for the
separation of Euphausidan larvae (at stages later than the Meta-
nauplius) from other Schizopodan and from Decapodan larvae.

The earliest Calyptopis-lsirvse captured by the ' Research '
resembled closely those figured by Sars {op. cit.) for other genera,
except for the facts that the carapace was much more globular
anteriorly and was devoid of spines or processes.

Nyctiphanes norvegica M. Sars.

This form was captured on only six occasions. Although a
North Atlantic type, it is not an essentially Arctic type like
Thysanoessa longicaxidata : it is of constant occui'rence in certain
localities on our own coasts, and has been recorded from as far
south as Portugal. The various records of its occurrence are cited
by Canon Norman *, but unfortunately the size of the individuals
and the depth from which they were derived are only rarely noted.
I am informed by Sir John Murray that, in his experience, large
adult specimens are taken only in deep water.

1 G. 0. Sars : Chall. Eep. Zool., xiii. Schizopoda, pi. xxvii. fig. 6.

2 C. Glaus : Untersucli. Crustaceen-Systems, pi. i. fig. 2, Wien, 1876, 4to.

3 With regard to these, Boas {oj). cit. p. 523, note 5) has suggested that they may
he homologous with the long caudal appendages of Nehalia and many Phyllopods.
This possibility is rendered considerably more remote by their being merely two
persistent spines out of a series which is not represented in the forms cited by him.

4 A. M. Norman : Ann. Mag. Nat. Hist. (6) ix. pp. 459^60 (1892).

1903.]

PLAMKTOX OF THE FAKROE OHAXNEL.

131

Text-fig. 13.

Text-fio- 15.

Text-fiff. 16.

Text-fio-. 17,

Telsoiis of l;ii'v:i' attribiitcd to Thiiaavocxsa Innf/icandald, sliowiiii;' vi'ductioii of tlic
primitive seven imirs of s|)iiies and torniatioii of tlie median uiijointed sjiine
of tlie telson.

Fig. 13. Early Calijptopis, 1'5 nini. long.

14. JPurcilia, 4"5 unn. long.

15. Ci/rfopia, 5 nun. long.

Ui. Late ('uvlopia, 5 mm. long.
17. 4 mm. long.

132 ON THE PLANKTON OP THE FAEROE CHANNEL. [Feb. 3,

The same distribution was noted by Vallentin and Cunning-
ham ^ : — " The adult, so far as our information allows of a
decision, lives on the bottom, and never swims far from the
ground [90-95 fathoms in this instance], while the young, up to
half or three-quarters the size of the adult, occur abundantly at
the very surface and at all intermediate depths. As mentioned
above, Mr. Murray found swarms of individuals at the surface in
the Faroe Channel, but none of these were full-grown, and very
few more than half the adult size."

I have no doubt that this generalization will prove true for
greater depths : I took adult specimens, over 35 mm. in length,
only between 350-220 fathoms, 400-300 fathoms, 500-400
fathoms: the remaining specimens varied from 9 to 17 mm. in
length. In other words, Nyctiplianes norvegica is apparently
mesoplanktonic when adult.

A few larvse, larger for their stage of development than those
attributed to Thysanoessa, were taken at the surface, and may
perhaps belong to JSfyctiphanes : they have not been included in
the tables.

Thysanopoda microphthalma Ortmann. (? = Thysanopoda
mia'ojihthaVma G. 0. Sars.) ,

Three specimens, recognizable as young forms by the character
of the second maxiUa and gills, and by the spination of the telson,
of about eight, twelve, and fourteen mm. in length, appear to be
referable to the same species as specimens recorded in quantity
by the ' National ' in 60° 3' N., 27° 0' W., at a probable depth of
between 218 and 328 fathoms. These were referred by Ortmann"
to Thysanopoda 'microphthahna of Sai'S ^, a species founded on two
specimens from the surface at 26° 21' N"., 33° 37' W., and 7° N.,
23° W. respectively. The identity of the ' National ' specimens
with those described by Sai'S seems to me rathei- doubtful : firstly,
because it is not very likely that a rare form svich as this should
occur as adult both at the surface near the Equator and also at
218-328 fathoms in the Greenland Sea, or at 500-400 fathoms
in the yet colder water of the Faeroe Channel ; secondly, because
Ortmann himself indicates some points of difierence between his
specimens and those of Sars. My own specimens agree with
Ortmann's figure, and dififer from Sars's description, in the shape
of the antennal scale, and in the absence of a spine from the
second joint of the first antenna. The telson was not hispid,
probably owing to immaturity ; the eye was somewhat flatter than
in Ortmann's figure, and showed slight signs of a constriction
such as is charactei'istic of Thysanoessa. The matter cannot be
settled in default of further specimens, owing to the fact that
Sars gave only a woodcut of the entire animal, and no figures of
the detailed anatomy.

1 R. Vallentin & J. T. Cunniugliam : Quart. Journ. Micr. Sci. xxviii. pp. 325-6.
- A. Ortmann : ' Decapoden und Schizopoden der Plankton-Expedition,' p. 9.
3 G. 0. Sars : Chall. Rep. Zool., xiii. Schizopoda, p. 106.

1903.] ON THE ELK IN NORWAY. 133

Of the tliree ' Research ' specimens of this species, two came
from the Mesoplaiikton, one from a haul of 480 to fathoms :
no examples were captured at the surface, unless some of the
larvae attril)utefl to 7"/^ ?/,s-rtv(oessa belonged to this species ; this is
imlikcly, liecause in tliat case the small size of the eyes would
prol);il)ly have l)etrayed them.

[). On the Present Condition and Habits o£ the Elk in
Norway. By H. J. Elwes, F.R.S.

fKeceivcd January 19, 1903.]

(Text-figures 18-26.)

tSo little seems to be known l)y naturalists concei'ning the actual
condition of this remarkable animal, that I think some of the
observations I have made dui-ing six Septembeis, which I have
spent entirely in the fascinating spoit of Elk-hunting, may be
worth recording in the Proceedings of this Society.

My experience has been gained entii-ely in the pro^'inces of
North and South Ti'ondhjem, where the Elk is more numerous
than pci'haps in any other pait of Europe, and where it seems to
attain a. gi'eater size and vigour, if one may judge liy the develop-
ment of the horns (see text-fig. 18, p. 134) than anywhere else in
Scandinavia.

I have also derived much vahial)le infoi'mation fi'oui Capt. Geiurd
Ferrand, who was, I think, the first Englishman to liunt regukxrly
in the same districts, and whose experience continued ahuost
witliout a break for 20 years fi'om 1865. This has been invaluable
in confii-ming, and to some extent modifying, my own observations
and what I have been able to learn from the natives of these
provinces.

Foi'ty years ago the Elk was a much scarcer animal in Ceutiul
and Northern Norway than it is now, or i-athei" was ten years ago,
since whicli time it has probaljly decreased in numbers. At that
time it was hunted for its meat alone by the proprietors of tlie
regions where it occurred, and in the lai-ge tracts of forest
belonging to the Government it was not allowed to be shot until
1880. Up till that time very few foreigners knew what magni-
ficent sport Elk-hunting was, and as the natives preferred the
meat of cows to that of bulls, which, after the rutting-season
begins, is extremely rank and hard, the bulls increased in numbers
and were able to grow to a size and age which they rarely
attain now.

The legal season for killing Elk was formerly much longer
than it is now, and in North Trondhjem lasted for three months.
Then it was cut down to 6 weeks, from the 1st September till the
15th October, and now has been still further reduced to a, month
or 3 weeks in some j^rovinces, Avhilst in Sweden only 15 days are
allowed during which the animals can legally bo killed.

134

MR. H. J. ELWES ON

[Feb. 3,

2 03

i2i n

Jzi p

^^
'5 2

OS'S

1903.]

THE ELK IN NORWAY.

Text-fiff. 19.

135

Horns of fully adult Elk killed in Bjoindal, 65° N., Sept. 4 1895 ; 18 points; a fail-
average size for this district, but uuich larger than usuallj' seen in the south,
(y'l nat. size.)

Text-fig. 20.

^^

Horns of a full-grown but not adult Elk, killed Sept. 1896 in Upper Namdalen, 65° N.
Supposed by natives to be 4 or 5 years old. Probably this type would develop
into such antlers as shown in text-tig. 18 if conditions were favourable,
(ji nat. size.)

Text-fig. 21.

of young Elk, sui)]>oscd to be 3 years old. These would probably develop
into such a head as that shown in text-fig. 19. (yf uat. size.)

136 MR. H. J. ELWES ON [Feb. 3,

This rapid increase of an animal which, on account of its great
size and conspicuous tracks, cannot escape unnoticed in an
inhabited country, must be entu-ely attributed to the wise game-
laws made by the Norwegian Government, and in most parts
honestly observed by the most law-abiding and well-governed
people I have ever met with. In former times the Elk, in Scandi-
navia, as in ISTorth America, was hunted down in winter on ski
(the Noi'wegian form of snowshoe), and slaughtered for its meat
by every peasant farmer, till it had almost been exterminated.
When, however, a law was made that it could be hunted only in
the month of September, which period was for a time somewhat
extended in North Trondhjem, its numbers soon increased, and
about 20 years ago attained such proportions that English and
German sportsmen began to visit Norway to hunt Elk. The
landownei's in some districts then discovered that, instead of
hunting themselves oi- paying Swedish hunters, who, from long
experience, were more expert, to kill their Elk for them, the
right of killing Elk, which is limited to one animal on each farm,
had a letting value ; and when many such rights are united so
that a large tract of country can be reserved to the lessee, this
value was worth a little trouble to maintain. The consequence
has been that, though poaching and killing Elk out of season is
not entirely unknown, yet it cannot be carried on extensively ; and
I have little doubt that the Elk will continue to thrive wherever
the country is suitable.

The statistics which I append show the numbei's of bull and cow
Elk which are known to have been legally killed in the various
" Amts " or provinces of Norway in 1894, and the average for the
previous five years, and may be taken as very nearly exact, though,
according to some, the returns for the southern provinces a,i-e not
so accurate as for the northern ones, and this record takes no
account of those illegally killed. Tt will be seen that only four
provinces of Norway (excluding Finmark), namely, Stavanger,
North and South Bergenhus, and Romsdal, all of which are on
the south-west coast and exposed to the warm and wet influence of
the Gulf-Stream, are without Elk ; and it may be added that in
those four provinces wild Reindeer are most numerous.

North Trondhjem is before all the rest in numbers ; and in this
province I believe the size of the horns is or was also much larger
on the average than in any of the southern districts (text- fig. 18,
p. 134).

It is said that the Elk is gradually extending its range north-
wards, and has appeai-ed in the southern parts of the province of
Nordland only in the last few years ; and there seems to be no
i-eason why it should not go still further, as in North America the
Moose (which is so nearly allied to the Elk, that I do not think
it can be looked on as more than a variety of that animal) is
found in regions where the climate and food is certainly not more
favourable to its habits than they seem to be in some parts of
Nordland and Finmarken.

The greater part of North Trondhjem and a large proportion of

1903.] rut; klk ix norway. 137

South Trondhjem are mountainous, barreu, and thinly populated
except on the coast and in some inland valley.s and fiords, and
are covei'ed with forests of birch, spruce, and Scotch la- up to an
elevation of about 2000 feet, wherever the ground is not too rocky
or swampy for these ti'ees to grow. Above that elevation there
are fjelds or bare mountainous ujjlands, the lower slopes and
sheltered dells in which are more or less clothed with birch, willow,
mountain-ash, and alder, mixed with stunted firs. There are
many large lakes and lai-ge are;is of peat-l)og ; but these bogs are
rarely so deep anil soft as to be impassable, and even after long
periods of rain a, man can cross them by picking his way.

The country is divided into so-called farms, most of which con-
sist of small patches of oats and potatoes, with from 10 to 30 acres
of mea,do\v-land which is mown for hay, and the produce of which
is eked out by small ricks which are put up wherever a, sledge-
load of grass can l)e got together in the forest. A lai-ge area of
forest and mountain is attached to ea,ch farm, and there are usually
one or two sietei-s ^ on the mountain, to which the cows are driven
for about three months in sunnner. Though most of these farms
belong to the occupiers there are some large private properties,
belonging to timber companies and public institutions, and nnich
of the higher fjeld and barren mountain remains in the hands of
the Government.

In the wilder and less populated districts the Elk ai'e hai'dly
disturbed during eleven months of the year except by the occ;isional
attacks of bears and wolves. Though the bear has l)ecome rarei'
of late years, the wolf, on the contrary, has appeared in distiicts
where it was until recently almost unknown ; but at present
they have confined their attacks rather to the semi-wild rein-
deer, which are kept in some numbers by the Lapps all along
the Swedish frontier, fi'om Roros northwai'd, and to the sheep, of
whicli every farmer has from 10 to 30. It seems to be doubtful
whether bears can kill full-grown Inill Elk", lait during my last
hunting-season I found the remains of no less than three Elk calves
which had been killed by them.

The favourite food of the Elk in summer consists almost entirely
of the twigs and leaves of birch, willow, and mountain-ash, and
in winter of the branches and bai-k of the same shrubs. Wherevei-
there is a grove of mountain-ash, Elk will live almost entirely on
it so long as they can get it, and in disti'icts where Elk ai'e ahun-
dant the ti'ee is constantly eaten down, so that it seems likely to
become much scarcer than at present.

Scotch-fir twigs are also largely consumed during the winter,
but do not seem to be eaten so long as mountain-ash can be
easily procured. Though I have never actually seen Elk editing
grass, I am assured by the native hunters and by Oapt. Ferrand
that they do so in sunnner to some extent, and also ])ite oft' the
fiower-heads of Epilobium and other plants. Water-lily roots,

* "Saeter" is the Norsk tcini for a shieling.

2 Capt. Ferrand has known a bull Elk beat oil' the attack of two bears in company.

138 MR. H. J. ELWES ON [Feb. 3,

which are said to be the favourite summer food of the Moose in
America, are also eaten, though the plant is rare in the districts
where I have hunted.

The Elk seems to spend the greater part of the day in feeding,
though it lies down for some hours to chew the cud. Some
hunters say that it has a regular time for lying down, and
will not hunt between 11 and 2 o'clock, because of the great
difficulty of approaching the animal when it is resting. I have,
however, seen them lying at all hours between 9 a.m. and 3 p.m.,
and have found them up and feeding at midday and later, so that
there is evidently no rule beyond the appetite of the individual
animal.

From its great size and the nature of its food, the Elk requires
a much larger extent of feeding-ground than any animal I am
acquainted with, except the elephant ; and I should suppose that
at least thi'ee square miles of suitable forest would not support
more than one Elk continuously, judging from the number I have
found in places where they were at home and undisturbed.

It is an animal of extremely solitary habits, and in summer
more than three will hardly ever be found together, and more
often only two. A family-party usually consists of a cow, a calf,
and a yearling ; very often the cow and the calf are alone, and
two bulls are frequently found together before the rvitting- season
begins. In winter, however, they are said to be somewhat more
gregarious, but I have never heard of more than nine actually
being seen in company in the autumn.

In some Swedish forests, which are strictly preserved for shooting,
it is said, however, that the Elk associates in larger herds, and has
become so numei'ous that much damage is done to the forest by
their biting off the shoots and tops of the young Scotch pines ; but
in such localities they do not attain the size and vigour that they
do whei-e they have a wider range and a greater choice of food, as
in the two pi-ovinces where I have observed them.

The Elk is commonly supposed to be essentially an inhabitant
of forests, and though this is to some extent the case, and in winter
it no doubt almost invariably keeps to the shelter of the forest,
yet I have lately become convinced that it is found on the higher
fjelds to a much greater extent than is generally known. As
high as the bii-ch is found, Elk may certainly be seen in summer ;
and the old solitary bulls in disti'icts where open fjelds exist
rarely descend into the pine-forest until the rutting-season begins.
From 1500 to 2500 feet is a very common range in summer, and
on the high mountains of Upper Tydal I have seen tracks much
higher than this, far out in the open f jeld on the Swedish frontier.
And in North Ti-ondhjem, where the timber-line is not so high, the
open hill- tops in some places are covered with Elk tracks and dung,
as though the animals had remained there dui'ing the whole of
the hot weather. Capt. Feri'and assures me that he has killed a
bull on the mountain above Lake Fsemen at an elevation of at least
3700 feet, on regular reindeer-ground, whei-e arctic willows were

1903.] THE ELK IN NORWAY. 139

the only shrubby plants. When found in such situations, they
may be stalked in exactly the same way a.s led deer, but when
disturlied usually make for the sheltei- of tlie forest ; and even on
the fjeld they nearly always lie in or clo.se to a small grove of
dwarf birch, and are in consequence not so e;xsy to find with the
telescope as red deer.

The Elk is supposed to be monogamous, and I have never seen
moi'e than one cow in company with a bull during the rutting-
senson ; but I am assured l)y a Lapp, whose knowledge of their
habits is very great, that in districts wheie cows are more
numerous than bulls, the older bulls will change their mates two
or thi'ee times diu'ing the season. This seems to be confirmed by
the fact, which 1 have often noticed, that at the end of Septend)er,
when the I'utting-season is approaching its height, cows are often
found without bulls, and the ti'acks of solitaiy Ijulls are found,
travelling presuma,l)ly in search of fresh cows.

Col. Walker, of Tykillen, Co. Wexfoixl, who has had great experi-
ence in Elk-hunting, relates the following event in a letter to me : —

"Thomas and I saw a magnificent bull with a splendid head in
company with a cow and calf ; they were quite in the open on the
edge of a small lake about half a mile from us.

" I noticed that whenevei' the bull went to the cow she ran at
him, and butted him very hard in the ribs ; he then each time ran
from hei', and browsed on the ti'ees. Thomas then told me that
it would be useless to stalk this bull as he would be gone in a few
minutes. His stoiy was that the bull only remains -with the cow
for three days, and then she lieats him oft', and he has to go and find
another. Thomas had hai'dly done telling me this when the bull
made his last appeal to the cow. She gave him a rough recejition,
and he at once stai'ted oft" at a fast trot. She remained grazing,
and we could seethel)ull going away very fast for nearly two miles
over a long stretch of open mountain, which took him aci'oss our
boundary. We never saw this bull again."

Another pi-oof of their polygamous habit is that on such ground
as that on which I hunted last year, whei-e duiing eight yeai's only
one cow has been killed for every five or six bulls, and where in
consequence the cows were far more numeious, I found no more
barren cows than in districts hunted by Noi'wegians only, where
more cows than bulls had been killed.

A point upon which I have never lieen able to get any certain
information in Norway is whether the sexes atti'act each other
by calling at night, as the American Moose do. The art of calling
is unknown to Scandinavian huntei's so far as I can learn, though
it is said to be practised in North Russia and Kurland. The
Aveather in NorAvay is usually so wet and stormy at the end of
September that thei'e is no inducement to lie out in the forest ;
and though I have spent many nights in remotely situated huts
and sjcteis, and have listened at dark and daylight, I have never
heard such a cjill as is made by the Moose. The cows do, however,
call to theii' calves, and I have heard bulls uttering a low grunting

140 MR. H. J. ELWES ON [Feb. 3,

noise when alone, and apparently looking for cows. On this
point, howevei', I quote Capt. FeiTand's expei'ience, which is that
he has on several occasions attracted bulls whose whereabouts
he knew, especially on a still fi-osty evening, by imitating the
grunting of the bull, which can be heard when they are looking
for cows at a distance of from 500 to 1000 yards, or possibly
moi-e. Moreover, he has heard the cows make a somewhat
similar noise, and the calves make a bleating noise when they
have lost their dam.

Col. Walker also writes to me on the subject of the cow calling
for the bull, as obsei'ved by him at Storvand, on the property of
Mr. Oollett near Mo, in North Trondhj em, in 1890 :—

" One day my hunter and I came on the spor of a bull Elk and
cow ; after some time we looked over a small hill and saw, in the
valley below it, the bull and cow about 150 yards from us. Seeing
that he was paying hei- great attention I waited to see what would
happen ; after one or two attempts he jumped on her, they having
their backs to me : I aimed at him, but waited until he had finished
his performance ; as he came ofi' her I fired, and shot him stone
dead. He rolled ofi' on to the gi'ound. She did not appear to
have noticed the shot or that he was dead, but began to graze
close to him, and once or twice went over and smelt him, then
grazed again. I then walked up to the bull ; she did not seem to
mind me, only stai-ed at me and stood close to the bull. When I
was within twenty yards of her she got alai'nied and went ofi",
very slowly at first, constantly looking round for him to follow
her. Shortly afterwards, while we were cutting up the bull, we
heard a very loud harsh roaring noise, just like the noise made by
a badly wounded bear when pinned in a corner. Both my hunter
and I thought it was a bear. I then went to the top of a small
hill, and on looking over saw the cow in the middle of a large open
bog. She was roaring, I could see her quite plainly thi'ough my
glasses ; she then got wind of me and bolted off.

" On another occasion I had killed a bull which was in company
with a cow and two calves ; these bolted, the cow one way, calves
the other. About 7.30 p.m., when I was in my hut, I heard a cow
calling just like a domestic animal in this country. I asked
Thomas whether we were near a farm ; he then told me it was
the Elk cow calling for her calves."

Though the scent and hearing of the Elk are unusually acvite,
their sight seems to be by no means so quick as that of deer : if it
were, the difiiculty of shooting them, which is already veiy great,
would be much increased. The usual method adopted in Norway is
to use a trained dog in a leash, which can scent the Elk at a distance
of a mile or more in the forest ; but owing to the difficulty of
approaching without noise in the thick forest, and the cunning
which the Elk almost invariably display in lying down so as to
get the wind of anyone following their tracks, only those who have
great patience, caution, perseverance, and an intimate knowledge
of the habits of the Elk and the ground, are successful in

1903.] THE ELK IX KORWAY. 141

getting- more than a glimpse of a frightened Elk disappearing
among the trees. But if you can succeed in seeing the Elk l){>fore
he has hea,i-d oi- smelt you, you may take lihei-ties with his eye-
sight which would not he possihlc in deerstalking. I once
appi'oached a cow Elk, which was lying with her calf on an open
hillside, within 10 yai-ds hy keej)ing her hody hetween me and
her head, and might have got nearer if the calf had not seen me.
On another occasion I rowed in a boat, just as it was getting
daylight, within 150 yards of a hull Elk, whii-h had hcon lying on
the shore of a lake, and had not yet got up from his laii-, and
shot him from tlie boat before he had taken alarni.

When bulls are rutting they sometimes run towards a man wliom
they have heard but not seen, thinking that it is another Elk, and
in conse(pience stories are told of their attacking human beings.
But these stories will rai-ely bear investigation, and though there
ai'e undoubted instances of wounded Elk attacking human beings,
I never but once saw one turn upon me, and then only because
he could not get away.

Besides the system of hunting tbem witli a- dog in a leash, they
are also hunted to some extent with loose dogs which bring them
to bay ; and though this system reipiires a man of great endurance
in order to follow fast enough to keep the hounds within hearing,
sometimes for many hours, it is perhajis more exciting and deadly
than still hunting. When Elk are so disturbeil by dogs and
have not seen, heai-d, or smelt man, they will sometimes come to
bay veiy soon, but more often I'un many miles, and ah\'ays tiy
to throw the hoinids and hunter out by crossing large rivers or
lakes. The Elk is an extremely sti-ong and good swimmei-, and
(piite at home in the watei-, and though he usually ciosses a large
river where it is shallow, he can, when pi-essed, pass a. I'oai-ing
torrent, whei-e no boat could live. In the veiy wet autumn of
1893 I lost several of the bulls I hunted, in consequence of their
ci-ossing i-ivers which we could neither foi'd nor swim ; when
severely wounded an Elk iisuall}' takes to water.

They may sometimes be driven successfully in places wlune the
groinid is very steep or confined, but as a, rule Elk do not follow
particular paths so much as most large animals, a.nd can go up and
down very steep I'avines anil rocky places which would be thouglit
impossible for so large and heavy an animal. They usually feed
u]) wind and I'un down wind, but there is no rule about this, and
after a long hunt they sometimes come back neai'ly to the same
ground they started from.

On the whole, I consider Elk among the most difiicult animals
to hunt that I have ever had experience of, and, even where they
are (piite numerous, Ihave more than once spent ten days w Ithout
getting a fair chance at an animal worth shooting.

The bulls are said by some to shed their horns as early as
Christmas and not later than Januaiy, as Mr. Meade- Waldo tells
me they have done in our Gardens. Capt. Ferrand, however,
luis been infonncd l)y woodcutters who were worl<iiii,r ;dl winter

142

MR. H. J. ELWES ON

[Feb. 3,

1903.]

THE ELK IN NORWAY.

143

1^

144 MR. H. J. ELWES ON [Feb. 3,

in the forest, that March is a more usual time, and this seems
more probable having regard to the time at which the horns of
other deer are shed. The new ones begin to grow in April
or May, and are fully developed about the middle of August. In
September the velvet is rubbed ofF against a young fir-tree,
which is usually destroyed in the process, and from the middle to
the end of September the rutting-season commences. At this
season the bulls have a very strong, rank, musky smell, and
scrape shallow round holes in the ground like stags' wallows, which
retain the scent of their urine for some days. Occasionally the
bulls fight desperately, and I have seen places where the ground
was torn up over many yards and sprinkled with hair and blood,
but few hunters have had the good foi'tune to witness such an
encounter.

The development of the horns of the Elk in the Nfiinsos district
of Norway is apparently greater than in South Noi'way and
Sweden, or in the districts which it still inhabits in East Prussia
and Kurland, though I cannot speak with the same certainty as
to Russia.

On my I'eturn from Siberia, four years ago, I saw semif ossil horns
of the Elk from the district of Perm, which were larger than
any I have seen in Europe ; but on visiting the best collections I
could hear of in St. Petersburg, I saw none which appeared to me
better than the best Norwegian heads, or so good as the one
figured (text-fig. 23, p. 143), which is that of a Lithuanian Elk
in the Branicki Museum at Warsaw.

The bull calf has no horns the first year. The second year he
has a small spike on each side, sometimes, but rarely, two spikes,
and I have seen one on one side and two on the other. As he
grows older it is supposed that the number of points increase
annually, one on each side ; but I think this is not at all invariably
the case, and that the size and strength of the individual has much
more influence than the a,ge, and probably, as in the case of other
deer which shed their horns annually, food and climate have a
good deal to do with it. As a rule the horns increase in size and
number of points up to 10 or 12 years old and possibly more, and
in old animals have a tendency to diminish in size and in the
length and strength of the points. From 18 to 22 points are
pei'haps the aveivage of adult males in Noi'thern Norway, and 10
to 14 points in Southern Norway.

The widest span which I have ever seen is 54 inches (text-
fig. 18, p. 134), and the greatest number of points 16 on one side,
as on the shed horn of which I show a tracing (text-fig. 22, p. 142).
The breadth of the palm in this instance and the numbei' of points
are quite exceptional, and I very much doubt whether there are
any such Elk now alive in Norway, as Herr Bruun of Trondhjem,
who has the opportunity of seeing all the finest specimens which
are procured, says that he has seen nothing equal to them.

Moose heads of as great or greater size are, however, often
killed, and though the largest which I have ever seen is 65 inches

1003.] THE ELK IN NORWAY. 145

iicrnss, J liave lieard on t'.iii-ly good ivutlioi-ity of ono ovov (\ foot in
s|):iii. (This was written before anytliinn' w.is known <>[ iJio
^i^'aiitie iioiuls found in Alaska.)

VN^itli regard to tlie comparative size .-uid woiglitof the KIk .ind
Moose, I cainiot speak positively, as it lias never been |)ossil)le to
weigh an entire animal in tlie places where I have killed them.

Ml-. Abel Chapman, however, givo^s me the following weights of
a bull, of fair but not unusual size, killed by him on Se[)t. liltl) : —

KilonT.aniincs. ICilosiraiiinics.

TToad with skin of neck 40 = 40

Shoulders 42 each = 84

Ha rinches with foot 54 „ =108

Si.los 40 „ = 80

Nock and back 60 = GO

Skin, say 30 = 30

Total 402

This would make the weight of the living animal at least
1000 pounds or over, as the intestines are enoi'mous.

Capt. Feriand's best bull, a lO-3'^ear old, weighed 1240 English
pounds witliout the intestines.

The pa,ce of Elk, when undisturbed, is a slow walk, and their
movements are very deliberate, but they can ti-ot for many
miles over boggy and I'ough ground at a pace of 6 or 8 miles an
hour, and occasionally when much frightened break into a
Inmbering canter.

The female Elk has her first calf at three years old in tlie month
of May, which malces the pei-iod of gesta,tion about 7 months.
Usually she has only one, but not unfi-equently two cahos, \\hich
grow very rapidly, and ))y the end of September are as big as a
red-deer hind. They suck their dams until late in the winter,
and keep company with them until another calf is dropped, and
sometimes longer. They have occasionally been tamed in Sweden
and taught to go in harness, but owing to the difficulty of feeding
them they are not easy to keej) in confinement. I ha^-e seen,
however, in the Zoological Gardens at Rotteitlam a, cow Moose
which is said to have been about 20 years in confinement, and
I hope that we shall be equally successful with the one which was
captured by Mr. Nickalls and presented by him to oui- Gardens.

Statistics of the numl)ei' of Elk kilk^d in Norway in the season
of 1894. (Translated from the ' Tidsskrift Norsk. Ja-ger og
Fisker Forenings,' Heft 3, 1895.)

Name of All! t. No. of P.iills. No. of Cows. Totnl.

Smaalenene 8 3 11

Akershus (w 60 127

Hedemarkou 74 72 146

Kristians 90 87 177

Prog. Zool. Soc— 1903, Vol. I. No. X. 10

146

MK. H. J. ELWES ON

[Feb. 3,

Name of Amt.

Buskerud

Jarlsberg og Larvik

Bratsberg

Nedenes ,

Lister og Mandal . ,
Sondre Trondhjems
Nordre Trondhjems

Nordland

Stavanger

Sondre Bergenhus .
Nordre Bergenhus .,

Romsdal

Finmarken

No. of Bulls. No. of Cows.

69
9

54
3
1

50

212

9

73

8
48

7

50

194

4

Total.

142

17

102

10

1

100

406

13

646 606 1252

Average for the 5 years, 1889-93. 610 512-6 1122-6

To give a better idea of the numbers of Elk found in what was
perhaps, at that time, the best forest in Norway, I give the
following statement copied from my diary in 1895. The number
actually seen may possibly represent from ^ to g of the number
actually on the ground at the time : —

Cows &

Bulls, yearlings. Calves.

Sept. 2. Haende 1 ... Barren cow shot for meat.

3. Storen ... 1 1

4. Strommen... ... 1 1

5. Strommen... 1 11 ... 1 bull killed : 18 points.

6. Blank day.

8. Mo Ill

9. Mo 1

10. Oplo 1 1

11. Syiinses ... 1

12. Lia Ill 1

13. Lia 11 ... ... 1 bull killed: 16 points.

15. Blank day.

16. Mo Ill

17. Storengen 1 1

18. Storengen ... 1 1 1

19. Fos 1 1

20. Mo 11 1

21. To Storvand, blank day.

22. Vennevik ... 1 11 1 1 bull killed: 6 points.

23. Synnffis 11 111 1 1

24. Brotten 11 ... ... 1 bull killed: 14 points.

25. Synnses 1 1 1 1 bixll killed : 13 points.

26. Quisten 1 ... ■■- 1 bull killed: 11 points.

27. Oplo 1 1

29. Mo Ill 1 bull killed : 18 points.

Total 17 25 14

This number of Elk was found on about 12 farms out of 24
over which I had the right of hunting. These farms might
average 10,000 acres more or less in extent.

On the same ground as this Col. Walker kept the following
record of the Elk which he saw himself during the four seasons
1887-1890, between Sept. 1st and Oct. 10th. It tends to show

1903.]

THE ELK IX NORWAY.

147

how rapidly Elk increase where tliey are protected and the cows
not killed : —

Year.

Old
Bulls.

Younsr
Bulls.

Cows.

Calves.

1887.
1888.
1889.
1890.

3
7
6
5

10

j 10
8 7

17 14

18 1 20

1

Two of tliosc cows hiul two ("lives each.
One of these cows had two calves.
Six of these cows had two calves each.

Nine of tliPS(! cows jiad two calves each.,

]

Most of these observations were wiitten at the reqnest of
onr late President, Sir W. Flower, six years ago, but were
withheld from pnlilication until this year, when the question
of a new species of Elk existing in Siberia was raised l)y Mr.
Lydekker and thejion. W. Rothschild. I then added the following
M})pendix.

Text-fio-. 24.

Morns of a very old Elk showing- deijeneration. (About ^V nat. size.)

Appendix. — On the Variation op the Elk.

J)v. Li)nnl)ei'g's papei' on the vai'iation of the Elk appears to
me to confirm the opinion I expi'essed at a recent meeting of the
Society (P. Z. S. 1902, vol. ii. p. 144), that the Siberian Elk
described by Mr. Lydekker (P. Z. S. 1902, vol. i. p. 207) as Alces
hedfordia' was nothing more than an inconst;int Viiriety of the
Eui'opean Elk. The Hon. Walter Rothschild has since expre.ssed
M contrary opinion, foi' which, however, I am not aware that he
has any fuither evidence than the fact that other horns similar

10*

148

MR. H J. ELWES ON

[Feb. 3,

to those of the so-called A. bedfordice have been received by him
from Sibeiia. I stated that the only horns of the Elk which I
succeeded in obtaining in the Altai Mountains in 1898, which
are now in the St. Petersburg Museum, were palmated in pre-
cisely the same manner as those of the Norwegian Elk, and others
which 1 have seen in Russian collections had all well-developed
palmation. Dr. Lbnnberg's figures go to show that the develop-
ment of the antlers in those parts of Sweden from which they
come is nothing like so fine as in the districts of North and South

Text-fig. 25.
A

Tracing of a cast horn found at Solem in Bangdal bj' Capt. Ferrand.
(About ~ nat. size.)

Trondhjem, where most of my hunting has been done. When I
first began Elk-hunting in Tydal, a mountain valley in South
Trondhjems-amt running up to the Swedish frontier, I should have
considered the horns figured by Lonnberg in fig. 3 as a fair repre-
sentation of an adult Elk in that district ; but I have seen them with
as many as eleven points on each horn, and believe that horns with
even more points have been obtained. Further north, in North

1903.]

THE ELK IN NORWAY.

149

Ti-ondhjems-amt, wliei-e the Elk a few yeai-s ago was exti-emely
numerous, mucli lai'gei- heads occurred, owing, I believe, to the much
greater quantity of mountain-;ish, which seems to be the favourite
winter food of the Elk, anel wliicli has been to a great extent
desti'oyed by the vast numbei' of animals constantly devoui'ing it.
Out of the whole numlier of bulls I have killed, about twenty,
only four oi' five had horns of the type shown in Lonnberg's

Text-fiL^ 26.

One-homed Elk in Ipswich Museum.

figui'es 1, 4, 6, and 9 ; and all of these, with one exception,
appeared to be quite young animals.

Among the huntei's who accompanied me in different years
were a Swede, a Norwegian, and a Lapp, all of whom knew the
Elk of those districts most intimately, and were considered the
most expei-ienced huntei's of theii- districts. None of them ever
even sviggested the possibility of two races of Elk existing ; and I

150 ON THE ELK IN NORWAY. [Feb. 3,

attribute non-pahnation of the horns entirely to impeifect
development of the animal, caused by insufficient food in their
earlier years, or degeneration caused by old age, wounds, or other-
wise. We did not reckon an Elk to be adult until he had at
least seven points on each horn, and the oldest and largest bull
that I ever killed, which had been well known in the district, and
continually hunted for at least ten years, had, when I killed him,
well-palmated hoi-ns of about one-half the size of what they ought
to have been in an adult animal of his size, showing that in the
Elk, as in the red deer, the horns degeneiute in size and number
of points in old age, which may be fifteen or twenty years or less.
Two such instances of degeneration are figured. Text-fig. 24,
p. 147, is the head of an old Elk killed by Thomas Bate, Esq., in
Lurudal Namdalen on Sept. 25,1 890. The horns measure 43 inches
in expanse, with ten points on one side and eleven on the othei- ;
the points are, however, not arranged in a uniform series on the
edge of the palm, and the development of the brow- antlers is very
abnormal.

The shed hoi'us of what was probably the same animal were
picked up on the same ground in the year previous by Col. Sullivan,
and ai'e of the same type with the same number of points.

An Elk displaying still more remarkable abnormal degeneration
was killed by Capt. Ferrand, and is shown in text-fig. 26, p. 149.
This animal was supposed to be 25 years old or more, aiid
the incidents of his death have been most graphically described in
the ' Badminton Magazine ' for March 1901 by Caj)t. Ferrand. The
horns are now in tlie Ipswich Museum.

The largest horns I have seen from Norway, belonging to an
animal which I unsuccessfully hunted for many days, but which
was afterwards killed by a farmer, and sold to me by Mr. Bruun
of Trondhjem, were 54 inches in width, with nine points on each
side (see text-fig. 18, p. 134); but there is a pair of shed horns in
Sir Henry Pottinger's house at Mo, of one of which I send a tiucing
(see text-fig. 22, p. 142), showing sixteen jooints on each side. It
is well known in Norway that the Elk of the southei'n districts,
which are much more fully timbered, and whei'e there is nothing
like the same extent of open fell and good feed as in North
and South Trondhjem, and where both sexes are much more
constantly hunted and the calves frequently deprived of their
mother's milk in Sej)tember, do not, in modern times at least,
produce anything like such fine heads as those of the wilder
districts of the north, the conditions being probably very similar to
those described by Di-. Lonnberg as in the southern provinces
of Sweden \ Taking his nine figures, I should be inclined to say
that all except 2 and 3 might, if they had come from Siberia,
have been considered as belonging to Alces bedfoj-dice ; and form,
to my mind, ample proof of that being (if the horns belong to

1 Mr. Percy Godman informs me that a well-known Elk-hunter in South Norway
considers that on his propertj' Elk have diminished by two-thirds in the last ten
years, and attributes this decrease to their having eaten and destroyed all the " leaf-
trees," i. e. willow, mountain-ash, and aspen.

P.Z.S. 1903, vol. 1. pi.xiy.

J.Picltarfl Gambridle (tfil etlith. West,NewiTian imp,

SPIDERS OF THE FAMILIES PISAURIDJi, AND SENOGULID^.

P.Z .S. 1903. vol.1. PI. XV.

^""""'^ 17.
F. Pic:Kard.Ca.mbri(ige.(lel.tfditii. WG5t,Newman imp.

SPIDERS OF THE FAMILY PISAURIDJi;.

1903.] MR. F. PICKARD-CAMBRIDGE ON NEW SPIDERS. 151

adult animals) iiotliiiig more than a local variety produced by com-
pai'atively unfavourable conditions. In consideiing the prol^ability
of the existence of local races of Elk, it must be remembered
that no northern mammal, except peihaps the reindeer, has such
wandering habits ; and though a race might })ecome temporarily
isolated in an area, which was entirely sui-rounded by a lai'ge
extent of country providing no suitable food, yet there is no
doubt, in my mind, that when their favourite food had become
partially exhausted, they would migrate many hundreds of miles,
and thus pi'event the establishment of local races.

A very interesting piper on the former existence of the Elk in
the Thames Valley in England, with a plate showing the great
similarity of its horns to those now existing, is j^iiblished by
Mr. E. T. Newton, F.Ii.S., in the Quarterly Journal of the
Geological .Society, vol. lix. (1903), and I am much indebted to
that gentleman for sending it to me.

February 17, 1903.

Dr. Henry Woodward, F.R.S., Vice-President,
in the Chaii-.

]Mr. R. E. Holding exhibited and made remarks upon an adult
skull of a Collie Dog, indicating a displacement of the incisors
caused by the closing of the left lower canine upon the third
incisor of upper jaw, and which apparently retarded the develop-
ment of the second incisor of that side. The skull also showed a
supernumerary canine which could in no way be mistaken for a
retained milk-canine.

Mr. Holding also exhibited portions of three Rabbits' skulls
having "over-shot" incisor teeth. In one specimen the abnor-
mality had caused a deviation fi'om the median line of the anterior
portion of the skull ; and in another the incisors, after leaving the
pi-emaxilla, had formed a complete circle and grown into the palate,
causing stai'vation and death of the animal.

The following papers were read : —

1. On some new Species o£ Spiders belonging to the
Families Pimuridcc and Senoculichc ; with Characters of
a new Genus. By Frederick Pickard-Cambridge,
B.A., F.Z.S.

[llccciveil Jamiary 10, 1903.]

(Plates XIV. & XV.')

While working out the Spiders of the two families Pisauridm
and HenocuUdai of Central America for the ' Biologia,' oppor-
tunity was kindly offered l)y Mr. R. I. Pocock for an examination
of the material in the British Museum Collection, with the result

1 For u.\plaiiatiou of the Plates, see p. 168.

152 MR. F. PICKARD-OAMBRIDGE ON NEW SPIDERS. [Feb. 17,

that a new genus and eight new species were found, several of
them occuiring amongst those collected by myself on the Lower
Amazons in 1895-6.

The Spiders belonging to the Fisaimdce are interesting from
the fact that they run freely and with great rapidity over the
surface of the watei", even in a sti'ong curi-ent. One of the genera,
jDolomedes, and another, Thalassius^ have both been dredited with
captiu'ing and devouring small fish of various kinds ; and of the
latter genus Mr. A. N. Stenning, himself formerly a gamekeeper,
whose observations are likel}^ to be trustworthy, declares that he
has found members actually devouring the small fry of a variety
of trout which occurs in South Afi-ica, some of the culprits being
now in the Museum Collection. Other notes on the habits of
these intei'esting Spiders will be fovmd under the genus Trechalea.

Species descrihed or figured heloiv.

Thaumasia velox B. Simon, 5 > P- 154, PL XIV. fig.' 1.

,, anmdipes, sp. nov., 2 ) P- 154, PI. XIV. fig. 2.

Dossenus marginatus E. Simon, 5 , p. 155, PI. XIV. figs. 3-5.
Paradossenus nig?'icans, sp. nov., $ , p. 155, PI. XIV. figs. 6-9.
Thanatidius spinipes, sp. nov., 5 , p. 156, PI. XIV. figs. 10-12.
/Senoculus pcmrdlelus E. Simon, $ , p. 166, PI. XIV. fig. 13.

cdbidus, sp. nov., 5 , p. 168, PI. XIV. fig. 14.
Trechcdea longitarsis 0. L. Koch, c5" 5 , p. 160, PI. XV. figs. 13-17.
„ keyserlingi^ sp. nov., 5 ? P- 163, PI. XV. figs. 1, 2.
,, urinator E. Simon, J $ , p. 161, PI. XV. figs. 3--5«.
,, ellacombei, sp. nov., 5 , p. 161, PI. XV. fig. 6.
,, macconnelli Poc, (^ , p. 162, PI. XV. figs. 7, 8.

coniiexa O. P.-Oambr., d , p. 162, PL XV. figs. 9, 10.

extensa 0. P.-Oambr., c^ , p. 162, PL XV. fig's. 11, 12.

,, amazonica, sp. nov., J $ , p. 163, PL XV. figs. 18-20.

Hesydrus pahistris E. Simon, (S $ , P- 165, PL XV. figs. 22-25.

hahilis O. P.-Oambr., S , p. 165, PL XV. fig. 21.

Fam. PisAURiD^.

/Synopsis of Genera decdt loith below.

A. Tibia i. with four (2-2-2-2) or five (2-2-2-2-2)
paired spines beneath, the last pair very small and
apical. Protarsi i. with three or four pairs beneath.
A'. Tibia i. with four pairs of spines beneath. Protarsi
i.-iv. with a single small central apical spine beneath.
Sternum produced into a long narrow conical point
between coxaj iv.
i. Tarsi shorter, not flexible. Protarsi i. with three
(2-2-2) paired spines beneath.
aa. Lower margin of fang-groove with 3 teeth.
a. Tibial and protarsal spines comjiaratively short,
only two or three times as long as the diameter
of the segment.
* Anterior row of eyes wider than that formed
by the posterior centrals. Central posterior
eyes nearer to each other than to the
laterals.
a^. Teeth on lower margin of fang-groove not

equidistant, the 3rd more remote Thaumasia Perty.

1903.]

MR,

F. nCKAUD-CAMBRIDGE ON NEW SPIDERS. 153

6'. Teuth on lower margin of fang-groovu

(■i|ui(list:int Tiniis V. I'.-Caml.r.

** Aiitfvior row of eyes not wider than that
fornii'd ])y the posterior centrals. Cen-
tral posterior eyes much further ai)art

than each is from tlie lateral Enna O. I'.-L'amljr.

b. Tiliial ami jirotarsal sjiiiies com])arative]y
very lon.y;, the lon;,'-est heini; ([uite two

thirds tlie length of the whole segment .. JJossemis E. Simon.
////. Lower margin of iang-groove with 4 teeth,

the 3rd smaller I'aradossenus, gc]\. nov.

ii. Tarsi long and ilexihle. Protarsi i. with four
(2-2-2-2) i)aired si)ines heneath.
a*. Leg.s sul)ei|ual in pairs, l — 2 and 1 — 3. .Man-
dibles conspicuously gibhous ahove, glahrous,
more distinctly earinate on the outer side
(males). Lower margin of fang-groove with

3 teeth Uesjjdrus Simon.

b*. Legs unequal, 4, 2, 1, 3. Mandibles not
conspicuously gibbous above, hairy, less
earinate (males). Lower margin of fang-
groove with 3, d, or 5 teeth Trccliaha Thorell.

H'. 'I'iliia i. with five jiairs of spines beneath. Pro-
tarsi i.-iv. with two apical spines beneath.
Sternum very broad between co.xa; iv., not

forming a point 'i TliauaLidius K. Simon.

1>. Tibia i. with 9-10 pairs of spines beneath, the last
pair not ai)ical. Protarsi i. with 8 or 9 pairs of
spines beneath Sijidrechalea F. P.-Camlir.

If the above genera he separated l)y the perhiips moi'e I'asily
observed character furnished hy the eyc-fornuda., then the tal)le
would run as follows : —

A. Anterior row of eyes not or only a little wider than the posterior

central row. r Enna

a. Anterior row of eyes not wider than the posterior central row... \ jjossenus
h. Anterior row of eyes a little, but distinctly, wider than the

posterior central row Faradossemis. Timis. Thaumnsia. lles,i/dri(s.

Trcchalea. Sijntrcchalca.

B. Anterior row of eyes much wider than the posterior central row,

as wide as the posterior lateral row Thanalldius.

Note. — (1) The number of protarsal sjiines in some species of Trechalea arc
difficult to locate, and the character is thus rendered less valunlile.

(2) One cannot be ([uitc certain that Thaii<((idiiis has lieen correctly identified.

(3) Tiiins is closely allied to Dranccs Simon, but the latter has, according to this
author, no apical pair of spines beneath tibia i. and ii., while they are present in the
former. The name Urauces (not Dranccs) was used by Champion for Coleoptcra
in 1889.

(1) In the l?iologia Centr.-Amer., Arach. Aran. ii. i). 312, I have referred the
species Hcsi/drus juUicni Simon to Enna 0. P.-Cambr. on the ground that they
have the characters, the tarsi flexible and central jjosterior eyes 2 diameters apart, in
common. This geims therefore comprises Simon's JJcsj/dnis Sect. 2 a, while
Hest/drus, as here diagnosed, comprises bis Jlcsi/drus Sect. la.

Genus Tiiaumasia Perty.
Del. Anini. Art. Bras. p. 192 (1833). Type, T. senilis Perty.

SjJecies recorded.

1 . T. senilis Perty, loc. cit Brazil.

2. T. nuinjinella (C. L. Koch) (Sub JJolomedes.)

154 MR. F. PICKARD- CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

3. T. sGO'paria (E. Simon) (Sub Scdtuinus.)

4. ? T. scajndaris (0. L. Koch) (Sub JDolomedes.)

5. ? T- binotata (0. L. Koch) (Sub Dolomedes.)

6. Thaumasia velox Simon, Ann. Soc. Ent. Belg. xlii. p. 18

(1898) Brazil.

Thaumasia velox Simon. (Plate XIY. fig. 1 .)

Loc. cit. Type S , in coll. E. Simon. Total length 8 mm.

Thaumasia velox F. P.-Cambr. Biol. Centr.-Amer., Arach. Ai-an.
ii. p. 309, t. 30. figs. 5, 5 a-c ( 6 )■

The type of this species {<S) was taken on the Amazons, and,
short of first-hand evidence, I do not doubt that the male and
female, taken by myself amongst the foliage on the margin of a
small " furo" in the Parana Buyassu, near Bi'eves, Lower Amazons,
must be referred to this species.

The male identified as belonging to this species from Guate-
mala may possibly be not identical, but it is impossible to be
certain with but one example from each locality.

The female has not been before described, but a figure of the
vulva is given on the Plate. For figures of the male see Biol.
Centr.-Amer. loc. cit. above.

Hah. Lower Amazons : Santarem ; Breves, Parana Buyassu
{F. P.-C, 1895-96); Rio Tocantins, S. Paulo de 01iven9a {de
Mathan) ; Guatemala (Sar-g).

7. Thaumasia annulipes, sp. nov., $ . (Plate XIV. fig. 2.)

Type 2 , in coll. Brit. Mus. Total length 5-5 mm.

5 . — Colour. Carapace deep bi'own, with pale yellow nai'row
central band, slightly dilate behind eyes. Abdomen brown,
shoulders and anterioi' lateral portion deep black. Legs pale
yellow, femora annulated with dusky brown.

Structure. Carapace horizontal above, convex and abruptly
deflected behind. Eyes : posterior row recurved, subequal, equi-
distant ; latei-als on a stout black tubercle. Anterior I'ow straight
or slightly procurved, wider than width occupied by jDosteiior cen-
trals. Anterior eyes equidistant, centrals larger. Clypeus more
than three tiraes the height of the diameter of anterior centrals.
Mandibles long and stout. Fang-groove with three denticles on both
margins. Sternum cordiform, posteiior angle attenuate, produced,
its apex extending as far as the pedicle. Coxce of pedipalp long,
straight, slightly dilate and i-ounded at apex. Lahmm half the
length of maxillfe, broad, squarely truncate at apex. Legs
4, 1, 2, 3 ; iii. the smallest, extending as far as apex of tibia iv.
Tibia i. and ii. with 2-2-2 spines beneath ; 1 — 1 lateral on apical
half ; 1 — 1 at base and apex above ; 2 at apex beneath. Pro-
tarsus i. and ii. with 2-2-2 spines beneath and 1 at apex ; 1 — 1
on each side ; none above. For vulva, see PI. XIY. fig. 2.

Several females of this small Spider were taken by myself in the
neighbourhood of Manaos, Lower Amazons, 1895-6.

1903.] MR. F. PICKARD-CAMBRIDGB ON NEW SPIDERS. 155

(lenus DossENUS Simon.

Hist. Nat. Ar. 2, ii. p. 314. Type, D. marylaatas E. Siuiuu.

Lowei" Amazons.

DossENUs MARcaxATtis E. Siiiion. (Plate XIV. figs. 3-5.)

Ann. Soc. Ent. Belg. xlii. 1898, p. 19. Type $ , and androtype
<S in coll. E. 8imon. Total length, 5 , 7"7 mm.

Two adult females of a Spider, wliich I feel pretty certain nnist
l»e i-efei'i-i'd to this species, wei'C taken by myself at lireves and
Parti, and are now in the British Museum Collection. The sj lines
on the legs are very long in compai'ison with those of allied
genera.

Hah. IsL. Tui.viDAL) ; BitAzir. : Ama/ons; Matto (Jrosso {Ger-
main); Lower Amazons, Pani, Breves (Parana Buyassu, /'. l\-<'.,
1895-96).

Paradossenus, gen. nov.

Type, P. nigricans, sp. n. Lower Amazons.

Lower margin of fang-groove with 4 teeth, the 3rd being
smaller. Tarsi shoit, straight, not flexible. Pi-otarsi i. with
3 paired spines beneath ; protarsi of all fom- pairs with a single
central apical spine beneath. Til)ia i. with four paired spines
beneath, the last pair apical. Sternum produced into a conical
point between coxaj iv. Spines on legs short, two to three times
the diameter of the segment. Posterior row of eyes recurved ;
eyes subequal, centi-als one diameter apart, two and a half from
the laterals, which are set on a low tuljercle. Anterior row
straight or very slightly procurved, slightly wider than the central
posterior row. Clyjieus one and a half times the height of the
diameter of an anterior central eye.

Legs 1, 2, 4, 3 — iii. very shoi't, extending as far as two-thirds
of tibia iv.

Paradossenus nigricans, sp. nov. (Plate XIV". figs. 6-9.)

Type S J 8 mm. ; gynetype $ , 8"5 mm. — carap. 3*75 mm.
In coll. Brit. Mus.

d 2 . — Colour. Carapace dull yellow-lirown, with two longi-
tudinal dark brown bars on each side of the pale central line, the
whole lieing almost entirely clothed with yellow-grey scale-like
hairs. Margins of cai'apace i-eticulated with fine dusky lines.
The anterior })ortion of the brown bars presents within their
margin a narrow elongate accent-like pale mark. Abdomen pale
yellow ; dorsal area with a broad longitudinal dark brown (some-
times paler) foliated band, becoming narrowed towards the
spinners, from the centre of whose lateral margin, on l)oth sides,
issues an ol)licpie interrupted lateral bar. 'I'he centre of the
foliated band is occupied by ,a pale lanceolate liand, much attenu-
ated towards the spinners, its basal portion again occupied by a
broad lanceolate dark band, scarcely I'eaehing the centre of the

156 MR. p. PICK ARD- CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

abdomen. Lateral ai-ea mottled with dusky brown ; ventral
sui'face pale yellow, unicolorous. Legs entirely yellow : i. and ii.
very faintly annulated and speckled with dusky brown, iii. and iv.
annulated, in some cases very richly, with black or brown.
Mandibles i-ed-yellow, with a large brown patch on each in front.

N.B. — These coloui's are very variable and in some cases the
central dark band within the inner pale band on the abdomen
is itself pale.

The vidva consists of two parallel plates, which have a promi-
nent angle on their inner margin. Between these, in front, lies
a deep transverse semicircular depression, and a convex central
piece, with a few shallow longitudinal grooves, is situated between
the plates.

For figui-es of tibial spur of male palpijs and the vulva, see
Plate Xiy. figs. 7, 9.

These spiders are very abundant on the Lower Amazons, racing
over the surface of the water with legs extended and body flat,
and are exceedingly difficult to catch. None of the females had
egg-cocoons with them, but they pi'obably cany them in the same
position as Trechalea, for in habits and general appearance these
spiders ai'e very similar to those belonging to the latter genus.

Hah. Lower Amazons: Breves, Parana Buyassu {F. P.-C,
1895-6).

Genus Thanatidius E. Simon.

Hist. Nat. Ar. 2, ii. p. 293. Type, T. dubius (Hentz).

N. America.

The following thi'ee species ai'e referi'ed to this genus by
Simon : —

1. T. dubius (Hentz), Bost. J. N. H. v. p. 449 (1847).

(Sub Thomisus.)

2. T. tcenius (Hentz), loc. cit (Sub Thomisus.)

3. T. imdulatios (Keys.), Yerh. z.-b. Ges. Wien, p. 486 (1887).

(Sub Tetragonophthalma.)

4. Thanatidius spinipes, sp. nov. (Plate XIV. figs. 10-12.)

Type 5 in coll. Brit. Mus. Total length 6 mm.

Hab. Parana Buyassu, Breves, Lower Amazons.

5 . — Colour. Carapace, mandibles, sternum, and legs bright,
unicolorous orange. Abdomen olive-green, with a central band
of cretaceovis- white pigment-cells.

Structure. Carapace nearly circular, ocular region produced,
parallel- sided, slightly raised. Thoracic region gibbous, abruptly
deflected behind. Eyes : posterior row strongly recurved, eyes
subequal, less than one diameter apart, equidistant ; laterals on a
stovit conical tubercle. Anterior row strongly procurved ; eye-
tubercles subequal, less than one diameter apart, equidistant, much
wider than the width occupied by lateral posteriors. Glypeus.,
in centre as high as four diameters of anterior central eyes.

1903.] MR. F. nCKARD-CAMBRIDGE OX NEW SPIDERS. 157

Mandibles long, slender, more than twice as long as lieiglit of
clypens. Famj-groove with tln-ee dentieiilos on lower uiaigiii.
MaxilliH long, parallel, dilate at apex; lal)imii Iiroad, liall' 1l:c
lengtli of maxilhe, ti'uneate at apex.

Sternum longer than broad, cordifonn, produced ])Ostoriorly
between coxai iv. into a hroad rounded j^lfote, the coxa^ being their
own diameter apart. Leys 1,4,2,3. Femora with two long
doi'sal spines and seveiul othei's on the sides and apex. Femnr i.
with two long spines in front in apical half. Ti])ia i. and ii. with
2-2-2-2-2 long spines beneath, the longest (the basal) being half
as long as the segment itself. 1 — 1 lateral spines ; 1 — 1 dorsal.
Protarsus i. and ii. with 2-2-2-2 long spines beneath (basal [)air
longest) : 1 — 1 lateral, 1 — 1 dorsal. Spines on tibia and pro-
tarsus iii. and iv. numerous, but less regulai-. Patella) 1-4 with
a single long apical spine. Scopula absent. Tarsal claws 3,
superiors with thi'ee or four denticules beneath. Spinners 6,
postei'iox'S longest. Pedipcdp with large tarsal claw, having 3
very long denticules beneath towai-ds l)ase. Vulva, see PI. XIY.
fig. 12. ■

Hab. A single 2 fi'oni the Paiuna Buyassu, near Breves, Lower
Amazons.

Genus Treciialea Thorell, 18C9 (nom. nov. for Triclaria).

Type, T. longitarsis (0. L. Koch), Colombia. Sub Triclaria
(1848), nom. pra>occ. by Wagler for Aves (1832).

The following characters are common to all the species men-
tioned below : — Teeth on uppei- margin of faiig-groove 3, on the
lower, 3, 4, or 5. Tarsi long and flexible. Tibia i. with four
pairs of spines beneath, the last pair small a,nd apical, besides two
latei'al spines. Protai-si i. with foui' paii-s of spines beneath (often
ii'regularly situated) ; protarsi of all foiu' pairs with a small
central apical spine beneath. Anterior I'ow of eyes slightly
I'ecurved, extending laterally slightly beyond the posterior
centi'als. Lateral antei-iors smaller than the central antei'iors.
Legs variable in length, 4, 2, 1, 3. Clypeus varialile in height,
porrect, and the eyes also vai'iable in their i-elative positions.

As regards the type-species of the genus, two distinct forms
have been identified as longitarsis of Koch — one by Keyserling,
now in coll. Biit. Mus., a female, having 5 teeth ; the other by
Simon (Ann. Soc. Ent. Belg. xlii. p. 20, 1898), male and female,
having 3 teeth. Since they cannot both be the true longitarsis,
I describe Keyserling's identification as a new species, for it is
totally distinct from examples from Colombia, now before me,
whence the t)/pe of the genus originally came. It forms, indeed,
the type of a, distinct grou]), so entirely difterent is tlie form nf
the vulva from that of the typical TrecJialea from Colombi.i.

At pi'esent, how'ever, although the identity of T. longiiarais
cannot be settled with absolute certainty until an examination
of the ty})e-specitii('n (if it slill exists in 1lii' Imperial Mnsenm at

158 MR. F. PIOKARD-CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

Berlin) has been made, the examples from Colombia are regarded
and here described as T. longitarsis C. L. Koch. All of the
species described and recorded below seem to be good and distinct,
though of course much more material from all parts is required
for comparison and the confirmation of characters.

The habits of these spiders are very interesting, for although
they are not amphibious in tlie true sense of the term, as are
Argyroneta and Desis, they are quite at home on the water, a.nd
speed along over the surface with great ease and rapidity, the
hairy clothing of the legs being waterproof. When racing away
over the water, they are not easy to distingviish from the large
Hemipterons (probably a species of Ranatra) of similar habit,
when the two creatures happen to be together in the same locality.
On being pursued over the water, the spiders will rush up the sides
of the boat and take refuge under the boards, but I have never
observed them dive below the surface as do Pirata, Dolomedes,
and some other Pisauridte. Whether, as has been reported of
certain species of Dolomedes and Thalassius, the spiders actually
prey upon small fish of various kinds, I cannot say from actual
experience, though the long flexible legs and exceedingly sharp
claws are well adapted for the piu'suit and capture of such pi'ey
in their native element. I may add that McOook's records of the
fish-catching capacity of Dolomedes (Amer. Spid. vol. i. p. 236 and
iii. p. 66) has recently received confirmation by Mr. A. N. Stenning
in South Africa. He tells us that Thalassius, a genus represent-
ing Dolomedes in the Ethiopian Region, has been often observed
by himself in the act of devouring the small fiy of a species of
trout, and calls the attention of pisciculturists in those regions to
the fact, and begs them to keep an eye on these spiders.

On a swampy island in the lai-ge largo opposite Santarem on
the Amazons, I have taken examples of one beautiful species
squatting flat, like Sparassids, on the truidvs of the trees, where
their hoary-white hairs and mottled legs and body afibrded them
excellent protective colouring. Nor do they resemble Heteropods
meiely when at rest, foi- when disturbed they dash round to the
opposite side of the tree-trunk with all the rapidity of a Selenops.
The egg-cocoon, as in many Pisaurids, is carried by the female
attached to the central pair of spinners at the tail-end of the
abdomen ; and it is cui'ious to note how, when these females stop
suddenly in headlong flight, the weight of the cocoon swings them
round on the watei-, and carries them along backwards for some
distance.

The species recorded and described below may be recognized by
the following charactei's : ^

Males.

A. Lower margin of fang-groove with 4 teeth, the third

smaller. Legs unicolorous urinator E. Simon.

1903.]

MR. F. PICKARD-CAMBRIDGE ON NEW SPIDERS. 159

B. Lower margin of f;inp:-jiroove with 3 subequal teeth.
Legs more or less si)otted or annulated.

I. Legs longer in ])roportion, leg iv. being about 8

times longer than the carapace.

a. Third tooth on U)wer margin of fang-groove more

remote from the others.

rt'. Tibia of pedipalp fromoue-half to twice longer

than broad ; tibial spur distinctlj' unculate at

apex, not concave on inner side. Embolus of

bulb with a long conspicuous spur at its liase. longitarsis C. L. Koeli
/>'. Tiliia of pedipalp not longer than broad ; tibial (sec. F. I'.-C).
spur not unculate at a])ex, concave on inner
side, the inner margin having a prominent
angle. Embolus of bulb without any long spur,

but having two chitinous ridges macconnelli Pocock.

h. Third tooth on lower margin of fang-groove not
remote, the 3 teeth almost e(|uidistant.
n-. Tibial spur not unculate at the apex.

Endjolus of bulb without any long spur connexa 0. P.-Cambr.

h'-. Tibial spur unculate at its apex. Embolus

of Inilb with a long spur at its base extensa O. P.-Cambr.

II. Legs shorter in proportion, leg iv. being 5-6 times

longer than the carapace amazoniea, sp. n.

An alternative table for Males.

A. Embolus of bull) with a long conspicuous erect black

spin- in its basal cavity.

a. Lower margin of fang-groove with 4 teeth m-inatur.

I). Lower margin of fang-groove with 3 teeth.
«!. Central posterior eyes one diameter apart ; apical
portion of embolus shorter, more distinctly
flanged on both sides ; 3rd tooth more remote ;

tibial spur less broad and flattened loiir/i/arsis.

hK Central posterior ej-es less than half a diameter
apart ; apical portio!i of embolus longer, less
conspicuously Hanged on the lower margin ; 3rd
tooth not remote ; tibial spur broader and more
flattened e.rfensa .

B. Embolus of bulb without any erect black spur in its

basal cavity.
a-. Tibial spur straight, blunt at apex.

rt'*. Tibial spur forming a cons])icuoas angle on its
inner maririn. Tooth 3 remote from tlie others;
apex of embolus less broadly flanged on its upper
margin maccauncUi.

/;■'. Tibial spur without angle on inner margin.
Teetli equidistant; apex of eml)olus more broadly

flanged on its upper margin connexa.

h'-. Tibial spur strongly curved, ]iointed at apex amazoniea.

Females.

A. Lower mavgin of fang-groove witli 5 teeth, the Ith

very small, situated on the inner side between the
3rd and otli. Vulva without any central tongue-like
sclerite, but with two small blunt cusps in tlie middle
of the anterior margin, and an obliq\ie elongate
lateral lobe on each side, converging behind tlie
cusps kpj/.ierli>i(ji. t<p. ii.

B. Lower margin] c)f fang-groove with 3 (or 4) teeth.

Vulva with a well-defined central tongue-like sclerite.
a. Lower margin of fang-groove with 4 teeth, the 3ril

smaller itriiui/ur E. Hhnuu.

160 MR. F. PICKARD-CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

h. Lower margin of fang-groove with 3 subeqnal teeth.

«'. Size much larger, 25 mm. Leg iv. at least ten

times longer than the carapace. Third tooth on

lower margin of the fang-groove slightly more

remote from the second than the latter is from

the first.

a". Central tongue-like sclerite much narrower,

dumb-bell-shaped, about eqtiallj' dilate at each

end ellacomhei, sp. n.

J-. Central tongue-like sclerite much broader,
almost as broad in the middle as posteriorly,
much broader anteriorly but somewhat vari

able longitarsis C. L. Koch.

h^. Size much smaller, 11 mm. Leg iv. not more
than 5-6 times longer than the carapace. Teeth
on lower margin of fang-groove equidistant.
Central cavity of vulva long and narrow, straight
on the inner margins ; central tongue elongate,
not dilate anteriorly, but narrowed amasonica, sp. n.

1. Trechalea longitarsis (C. L. Koch). (Plate XV. figs. 13-17.)

Die Arach. xv. p. 65, t. 522. fig. 1462 (sub Triclaria).

? Trechalea longitarsis E. Simon, Ann. Soc. Ent. Belg. xlii.
1898, p. 20.

Type $ , in coll. Mus. Imper. Berlin. Hah. Colombia.

Although the identity of this species cannot be satisfactorily
settled without reference to the tyjje-speci'men, if it still exists,
I give a desci'iption and figures of the characters of a form with
annulated legs which was taken in Colombia. If the type no
longer exists, and no more exact locality than " Colombia " can
be ascertained for the type- specimen, then the examples here
described stand as ^'- topoti/pes" — examples from the same locality
in which the type was taken — and will serve as a standard of
comparison for the species in the place of the original type.

Had tSimon given more definite characters (for the majority of
the species of Trechalea possess 3 teeth) which would enable us to
recognize his identification of T. longitarsis with any certainty,
and if these (for he makes no mention of the locality) were taken
in Colombia, then his examjdes would be the recognized " topo-
types " to which we should have to refer for an identification of
T. longitarsis (C. L. Koch).

Present identification of T. longitarsis (sec. F. P.-Cambr.).

Total length, S 20, ? 25 mm.— $ . Leg i. 61 ; ii. 68 ; iii. 55 ;
iv. 75 mm. Tib. i. 15 ; iv. 16"5 mm. Prot. iv. 21 mm.

Examples in coll. Brit. Mus. from Colombia.

(J 5 . — Colour. Carapace rich brown-black, with a pale orange
/V-shaped spot behind the eyes, having a central dusky line, and
one on each side ; a pale rufous-white, narrow, submarginal
band, and a narrow black velvety marginal band. Ahdomeyi
brown, with a black central foliated band : a more exact description
is impossible from dried and wrinkled examples. Sternum and
ventral area dull yellow-brown. Legs brown, very distinctly
annulated with rufous-white hairs, protarsi less distinctly, tarsi

1903.] MR. F. PICKARD-CAMBRIDGE ON NEW SPIDERS. IGl

not at all. Mandibles black, clothed with hoaiy-white hairs in
front.

Structure, General characters similar to those of other species of
the genus. Central posterior eyes one diameter apart (lass in S ) ;
clypeus six times as high as the diameter- of an anterior central
eye (in J four times).

Vulval area broader than long, rounded, rectangular ; central
tongue distally broad, widely dilate basally ; variable in diliereut
examples.

Palpus. Tarsus twice i\s long as broad ; tibia two-thii-ds the
length of the fcxrsus ; one-fourth longer than the patella. Tibial
spin- broad basally, curved apically, terminating in a small hook,
having on the inner side a slightly serrated ridge. The embolus
of the bulb has a stout erect spur springing from the cavity below
the apex of the lamina of the bulb, while the embolus itself is
short, strongly curved, and conspicuously flanged on each side.

Ilah. CoLOMBJA, dried examples ( c? »& $ ) ; Venezuela, dried
examples ( d" tfc $ ) ; also many examples in spirit, in coll. Brit.
Mus.

2. Trechalea urinator E. Simon. (Plate XV. figs. 3-5 a.)
Ann. Soc. Ent. Belg. xlii. p. 20 (1898).

Type S , gynetype $ , in coll. E. Simon. Total length ( c? , $ )
20-25 mm.— (c?)- Carap. 11. Leg i. 74; ii. 82; iii. 62; iv.
85 mm. Tib. i. & iv. 19-5 mm. Prot. iv. 22 mm.

Both sexes with 4 teeth on the lower margin of the fang-
groove. Legs 4, 2, 1, 3. Clypeus 3-3-1- times in height the
diameter of an anterior central eye.

Carapace, abdomen, and legs entirely without dark bands or
annules, or these are at least very faint ; entii'ely clothed A\'ith
dusty olive-bi'own hairs and pubescence.

Palpus. Tibia 1| times longer than broad ; tibial spur slender,
its apex bent over, forming a single small hook. The inner
apophysis deeply impressed across the middle, forming two blunt
cusps.

Vulva with a short broad central tongue, broadly dilate basally.

If I am correct in my identification of Simon's species, it is a
common Spider tlii'oughout Ecuador.

Hah. Ecuador : Guayaquil, Loja ; Rio Durango ; Bulun ;
Cachavi ; Cavondelet ; Paramba ; Salidero {Rosenhery).

3. Trechalea ellacombei, sp. nov. (Plate XV. fig. 6.)
Type $, in coll. Brit. Mus. Total length 25 mm. Cai-ap,

10 mm. Leg i. 68; ii. 75; iii. 60; iv. 80 mm. Tib. i. 17; iv.
17-5 mm. Prot. iv. 22-5 mm.

Lower margin of fang-gi-oove with 3 teeth. Legs 4, 2, 1 , 3.
Clypeus 4 times the height of the diameter of an anterior central
eye. The whole of the central area of the carapnce l)lack, with
margins and stria? flavous. Abdomen too much shrivelled to admit
of the coloration being recorded. Legs very deeply streaked mid

Proc. Zool. Soc— 1903, Vol. I. No. XI. 11

162 MR. F. PICKARD-CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

annulated with black, except the tarsi. Yulva with a long centi'al
tongue, dilate basally, and its apex incurved.

This species is nearly allied to T. i(,rinator, but may be recognized
by the annulated legs, the possession of only 3 teeth on the
mandible, and by the form of the vulva.

Hah. Surinam, Bergen Daal (collected by Mr. C. W. Ellacombe).

4. Trechalea macconnelli Pocock. (Plate XV. figs. 7, 8.)
Trans. Linn. Soc. Lond., Zool. (2) viii. 2, p. 67, Sept. 1900.
Type c? , in coll. Brit. Mus. Total length 16 mm. Carap.

9 mm. Leg i. 64; ii. 70; iii. 55"5 ; iv. 73. Tib. i, & iv. 17 mm.
Prot. iv. 18 mm.

Lower margin of fang-groove with 3 teeth. Legs 4, 2, 1, 3.
Clypeus twice the height of the diameter of an anterior central
eye. Carapace with a large black patch in the centre, its margins
scalloped by the flavous margin round the carapace and the pale
strife. Abdomen dusky with a flavous pattern and two flavous
spots above the spinners. Palpus : Tibia not, or scarcely, longer
than broad. Tibial spur slender, concave lamelliform, its lower
edge slightly serrate ; convex and incurved at the apex ; seen from
above it has a A-shaped emargination. The inner apophysis is
cylindrical, low and impressed across the top.

Sab. British Guiana : Mt. Roraima (collected by Messrs.
Quelch & McConnell).

5. Trechalea connexa (0. P.-Cambr.). (Plate XV. figs. 9, 10.)
Triclaria connexa 0. P.-Oambr. Biol. Centr.-Amer., Arach.

Aran. i. p. 233, t. 30. figs. l,la-e{S ), 1898.

Trechalea connexa F. P.-Cambr. loc. cit. ii. p. 313, t. 30. fig. 17
{6), 1902.

Type S , in coll. Godman & Salvin. Total length 14 mm.
Carap. 7 mm. Leg i. 40 ; ii. 45 ; iii. 39 ; iv. 48 mm. Tib. i. 10;
ii. 11 mm. Prot. iv. 13"5 mm.

Lower margin of fang-groove with 3 teeth. Legs 4, 2, 1, 3.
Clypeus as high as two diameters of an anterior central eye.
Carapace flavous, streaked and mottled with brown. Abdomen
dark brown. Legs decidedly annulate. Palpus : Tibia half longer
than broad ; spur, seen from below, without any hook at its apex.
Embolus of palpus without two ridges as in T. 'macconnelli, shorter
and more broadly flanged on its anterior margin.

Ilab. Mexico, Atoyac {R. H. Smith).

6. Trechalea EXTENSA (O. P.-Cambr.). (Plate XV. figs. 11, 12.)
Triclaria extensa 0. P.-Cambr. loc. cit. i. pp. 174, 233, t. 22.

figs. 10, I0a-f{s), 1896.

Trechalea extensa F. P.-Cambr. loc. cit. ii. p. 313, t. 30. figs. 16,
16«,6(c?), 1902.

Type S , in coll. Godman & Salvin. Total length 20 mm.
Carap. 10 mm. Leg i. 69 ; ii. 77 ; iii. 58 ; iv. 82 mm. Tib. i.
17; iv. 19 mm. Prot. iv. 20 mm.

1903.] MR. F. PirKARD-CAMBRIDOE ON NEW SPIDERS. 163

Lower margin of fang-groove with 3 teeth. Legs 4, 2, 1, 3.
Clypens 3 times as high as the diameter of an anterior centiul
eye. Carapace unicolorous flavous, margined with dusky black.
Abdomen unicolorous brown, with a short central anterior pale
dorsal bar. Legs faintly annulate. Palpus : Tibia twice as long
as broad ; spur, seen from above, stout, projecting, cnrved, with
a stout sharp cusp or hook on the inner side at the apex, very
similar to but much stouter than in T. imnator. Inner apophysis
deeply impressed in the middle. Embolus as in T. long liars is ,
but less bi'oadly flanged.

ITab. Guatemala {>Sarg).

7. Trechalea amazonica, sp. nov. (Plate XY. figs. 18-20.)

Type d", gynetype ? , in coll. Brit. Mus. Total length ^ 10;
( 2 15 mni.). Carap. 6 mm. Leg i. 31 ; ii. 32 ; iii. 29 ; iv. 37*5
mm. Tib. i. &, iv. 7-8 mm. Prot. i. & v. 10 mm.

Lower margin of fang-gi'oove with 3 teeth. Clypeus in height
equal to l^j diameters of an anterior central eye. Legs 4, 2, 1, 3.
Carapace dull stx'aw-3'ellow ; margins and sti'i;^ slightly suffused
with dusky brown, produced by fine dai'k hairs ; mottled with
patches of white hairs, angular on mai'gins ; clypeus fringed and
clothed with hoaiy- white hairs. Mandiljles, pedipalps, and a spot
on ocular area behind the eyes clothed with hoary- white haii's.
Abdomen olive-green mottled with brown and hoary- white.
Ventral area clothed with silver- white hairs. Sternum and legs
dull yellow beneath, the latter brighter above ; femora indistinctly
annulated with dusky brown above ; base of patella black. Tibite
and pi'otai'si with two broad dark annulations. Pedipalps also
annulated with brown.

Vvira with a narrow oblong central cavity, its lateral margins
straight and parallel, with an elongate central tongue, not dilated
at either end, not incurved at apes, without any black shining
bosses on each side.

The male and female here described wei-e taken from the same
trunk of a tree on a swampy island in the largo opposite tSautarem.
They were ci-ouching on the bark lilce Spai'assitls, and when dis-
turbed moved with great rapidity.

Hah. Lower Amazons : Santarem ; Breves, Parana Buyassu ;
Pard {F. P.-C. 1895-6).

8. Trechalea keyserlingi, sp. nov. (Plate XY. figs. 1, 2.)

Type $, in coll. Keys. Brit. Mus. Total length 20 mm.
Carapace 9'75 mm. Leg i. 44; ii. 46; iii. 36; iv. 49 mm.
Tib. i. 11 ; ii- 11 "5 mm. Prot. iv. 13 mm.

Lower mai-gin of fang-groove with 5 teeth. Clypeus five times
as high as the diametei' of an anterior centi'al eye. Legs 4, 2,
1, 3. Carapace with two broad centi'al bands, widely separate
behind the eyes, connivent behind ; tlie clear space l)ehind Hie
eyes marked with lateral and central fine streaks. Abdomen with
a broad deep-brown foliate band above, consti'icted just behind

11*

164 MR. F. PIOKARD- CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

the middle and again slightly above the spinners, broadly margined
with pure white pubescence. Femora of legs deeply blotched and
streaked above with black ; tibise distinctly, protarsi indistinctly
annulated. Vtdva without any central tongue, but with two
small cusps in the middle and two separate posteriorly converging
lateral sclerites, quite distinct in form from that of any other
species here recorded (see Plate XV. fig. 2).

This specimen M^as identified by Keyserling as T. longitarsis
C. L. Koch. See p. 157.

Hah. Brazil, Rio Grande do Sul {Keys. Coll.).

Genus Hesydrus Simon, 1898 (June 30).

Simon, Hist. Nat. Ar. 2, ii. p. 315. Type, H.'pahistris Simon.

Ecuador.

The species belonging to this genus are very closely allied to
Trechalea, but foim a small group chiefly distingu^ishable by the
glabrous, gibbous mandibles, with a well-marked carina on the
outer side, in the male sex. The legs in both sexes are subequal
in pairs 4-2, 1-3 ; being also much shorter in proportion than in
Trechalea. Both margins of the fang-groove with 3 teeth. Eyes
as in Trechalea^ central posteriors slightly over one diameter
apart ; clypeus low and porrected. Tibia and protarsus i. with
2-2-2-2 spines beneath; protarsi 1-4 with a small central apical
spine beneath.

In Biol. Oentr.-Amer., Arach. Aran. ii. p. 305, I have referred
the species given as the tyjoe of Hesydrus (sec. 2 a, H. jidlieni)
to Enna O. P.-Cambr., and the name Hesydrus is ajDplicable to the
species under sec. \a with H. jyahistris as the type. It appears
that Simon has somehow confused the characters of these two
groups. In Enna (including H.jibllieni) undoubtedly the clypeus
is moi-e vertical ; the centi-al posterior eyes are two diameters
apart, or moi'e, and the anterior row is not wider than the
posterior central row ; while the tarsi are straight and not flexible.
But in H. palustris (for if not identical, the form before me
is certainly congeneric with this species) the clypeus is pori'ect,
the central posterior eyes one diameter apart, the antei'ior row
wider than the posterior central, and the tarsi flexible ; the eyes
of the posterior row are certainly not equidistant, a character
given for distinguishing Hesydrus from Trechalea. Fortunately,
however, the citation of a definite type species enables one to
rectify the confusion.

The two species known to me may be recognized by the
following charactei's : —

a. Tibial spur of palpus, seen from above its apex, not bifid,

slightly emarginate and prominent at its angles, but

not incurving in the form of two teeth habilis 0. P.-Cambr.

b. Tibial spur of palpus, seen from above its apex, deeply

bifid, its angles forming two sharp teeth curving in-
wards and downwards palustris Simon.

1903.] MR. F. PICKARD-CAMBRIDGE ON NEW SPIDERS. 165

1. Hesydrus talustris Simon. (Plate XV. figs. 22-25.)
Ann. Soc. Ent. Belg. xlii. 1898, p. 21 ( J , $ ) ^

Type d'jgynetype $, in coll. E. Simon. Total length cj" 11
($16 mm.). — Caiap. 6 nnn. Leg i. 25; ii. 30; iii. 24-5;
iv. 30 mm. ' Tib, i. 6 ; iv. 7 mm. Prot. iv. 10-5 mm.

Lower margin of fang- groove with 3 teeth. Legs 4-2, 1-3.
Clypeus porrect, as high only as the diameter of iin anterior
fu>ntral eye. Cai'apace dull flavoiis with dusky haii's. Falpas :
Til)ial spur short, concave lamelliforni, deeply bifid at its apex,
forming two small, sharp, incurving teeth. Inner apophysis elon-
gate, not impressed across the middle, produced at its untci-ior
apical angle to foiiu a small cusp. Vulva with a broad central
cavity, more than two-thirds the transverse width, concave on its
lateral margins, having i\. long nai-row central tongue, not dilate
at either end, with a shiny black boss on each side, on the lloor
of the cavity ; very vai'iable in foi-m.

Hah. Ecuador: Vta-Ainh-A^Roseuhery). Loja, Zamora\ — Vene-
zuela : Merida.

It is not certain that my identification of this species is correct,
but it is highly probable.

2. Hesydrus habilis (0. P.-Oambr.). (Plate XV. fig. 21.)
Tridaria habilis O. P.-Cambr. loc. cit. i. pp. 173, 233 t 22

figs. 9, 9«-/(c?), 1896.

Trechalea habilis F. P.-Cambr. loc. cit. ii. p. 313, t. 30 fig 15
(c?), 1902.

Type S in coll. Godman and Salvin. Total length 10 mm. Carap.
5 nun. Tib. i. 5-5 mm. (other legs broken off or mutilated).

Lower margin of fang-groove with 3 teeth. Logs 2, 4, 1, 3
(sec. O. P.-C). Clypeus as high as 1| diameter of "an anterior
central eye. Legs dull yellow-brown, slightly annulate. The
colours of the cai^apace and abdomen appear to have faded.
Mandibles glabrous shiny and very gibbous dorsally. Palpus :
Tibia one-third longer than broad. Ti])ial spur, seen from above
its apex, not bifid, its angles not forming sharp incurving teeth.

Hah. Guatemala: Costa Rica (/S'cw*/, iioyers).

3. Hesydrus estabanensis E. Simon.

Ann. Soc. Ent. Belg. xlii. 1898, p. 20 ( J , ? ).

Type S , gynetype $ , in coll. E. Simon. Total length" J $ , 7 mm.

One would susjiect, from the descriptions, that this species is
congenei'ic with //. palusiris, the males having the luaudililes
convex and cai-inate on their outer apical margin.

Hah. Venezuela : San Esteban.

4. Hesydrus bucculentus E. Simon.
Ann. Soc. Ent. Belg. xlii. 1898, p. 20 ( c5' ).
Type S , in coll. E. Simon. Total length 10 mm.

The remarks made under the last species apply also to this.
Hab. Brazil : Rio {Gounelle).

165 MR, F. PICK ARD- CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

Fam. SenoculiDjE.
Genus Senoculus Taczanowski.

1872. jSenoGiohcs Tacz. Hor. Soc, Ent. Ross. ix. p. 106. — Type,

aS'. maronicids.

1873. Lahdams 0. P.-Cambr. Proc. ZooL Soc. Loncl. p. 118.—

Type, L. monastoides.
1880. Platyctenihs Keys. Yerh. z.-b. Ges. Wien, sxix. p. 338,
note. Nom. nov. for Senocidus (nom. inappro. sec. Keys.).
1880. Stenoctemis Keys. Yerh. z.-b. Ges. Wien, xxix. p. 340, —

Type, S. gracilis.
1883. Neothereutes Holmb. Bol. Acad. Nac. Cienc. Cordoba, v.

p. 35. — Type, N. darwini Holmb.
In a former paper, " Cteniform Spiders from the Lower Amazons
&c.," Ann. Mag. Nat. Hist. (6) xix. p. 90, Jan. 1897, 1 have tabu-
lated the characters of the known species of this genus so far as
I could diuw reliable characters from the existing descriptions.

The table is here re-modelled and includes additional species
from Central America, piiblished in the ' Biologia Centrali-
Amei'icana,' and a new species occurring in the British Museum
Collection, of which a diagnosis is given below. I have not seen
S. 2)ci,rallehts Simon, but it is certain that my example from the
Lower Amazons belongs to that species, since M. Simon has kindly
compared a figure of the vulva with that of his own type.

The following is a list of the species already published : —

1872. S. maronicus Tacz,, 5 , St. Laurent de Maroni. Hor. Soc.

Ent. Ross. ix. p. 106, t. iii. fig. 4.

1873. S. monastoides (0. P.-Oambr.), 5 , Brazil, Rio Grande.

P.Z. S. Lond. p. 118.

1879. S . riohromacidatus Keys., $, Peru. Yerh. z.-b. Ges.

Wien, p. 339, pi. iv. fig. 30.

1880. S. pl'umosus (E. Sim.), $ , Brazil, Para. Bull. Soc. Zool.

Fr. p. 154.
1880. >S'. pttrjnireus (E. Sim.), $ , Panama. Bull. Soc. Zool. Fr.

p. 155.
1880. S. paraUelus (E. Sim.), $ , Brazil, Tefife. Bull. Soc. Zool.

Fr.p. 156.
1880. S . riificapillus (E. Sim.), c? , Brazil, Para. Bull. Soc.

Zool. Fr. p. 154.
1880. .S'. iricolor (E. Sim.), S $ , Brazil, Tefi'e. Bull. Soc. Zool.

Fr. p. 153.
1880. S. gracilis (Keys.), $ , Peru, Aimable Maria. Yerh. z.-b.

Ges. Wien, p. 341, t. iv. fig. 29.
1883. S. dartoini (Holmb.), $, Formosa, Chaco, Arg. Rep.

Bol. Acad. Nac. Cienc. Cordoba, v. p. 35.

1896. S.2^rolatus (0. P.-Cambr.), S 2, Mexico, Atoyac. Biol.

Centr.-Amer., Arach. Aran. i. p. 218,t. 28. figs. 3, 3a-f{ $ ).

1897. tS. pallidus (F. P.-Cambr.), S , Brazil, Rio Janeii-o. Ann.

Mag. Nat. Hist. (6) xix. p. 92, pi. iv. fig. 5 (sub
iS'tenoctenus).

1903.]

MR. F. riCKARD-CAMBRIDGE OX XEW SPIDERS. 167

1902. S. canal icalatus F. P.-Cambr., J $ , Paiiaiiia, Biigalia.
Biol. Oentr.-Amer., Aracli. Aiun. ii. p. 350, t. 33.
figs. 3(c^), 4, 4«($).
The characters by which the species may be distinguished, so
far as I can gather them from the descriptions, ai-e as I'dHows : —

Males.

A. Carapace and abdomen clothed with plumose hairs, iricolor Siiiioii.

B. Carai)ace and al)domeu clothed witli simple hairs.

I. Tibia i. and ii. with 9 — 10, protarsi with 8 — 9 long

siiiiius beneath on each side j)alUdus F. I'.-C'auilir.

II. Tibia i. and ii. with 4 — i long spines beneath, pro-

tarsi with o — 5.

a. Anterior portion of ocular area almost vortical . ruJlcapiUus Siuion.

b. Anterior portion of ocular area porrcet, liori-

zontal.

1. Apex of lamina round tlio inner anierinr por-

tion of the bulb of the jmlpus nmeli broader

and more deeply and broadly bifurculato ... in'olatus (_). I'.-Cainbr.

2. Aj)ex of lamina round the inner ant('ri<ir

portion of the bulb of tlie palpus much
narrower and less deeply bifurculate (simply
bitid and canaliculate) canalicidatus F. l'.-(.'ambr.

Females.

A. Carapace and abdomen clothed witli plumo.so hairs.

1. Carapace deeply striate, central stria oval-elon-

gate; tarsi i. and ii. with 4 — i, protarsi i. and ii.

with 3 — 3 long spines beneath ■plamosiis Simon.

2. Carapace scarcely striate, central stria short, sub-

punctiform; tibia i. and ii. with 5 — o, protarsi

i. and ii. with 4 — 4 long spines beneath iricolor Simon.

B. Carapace and abdomen clothed with simple hairs.

a. Legs very long ; tibia i. and ii. with long spines

beneath, 10 on inner side, 9 on outer side.
«'. The lateral lolies of the vulva bearing a rounded

boss on the outer margin just behind the

middle (see Keyserling's tigiire) ffracilis Keyserling.

b^. The lateral lobes of the vulva without any

rounded boss parallchis Simon.

b. Legs sliorter, tibia i. and ii. with only 4 — 1- or

5 — 5 long spines beneath.

aa. Tibia i. and ii. with 5 — 5 long spines beneath.

(i'. Posterior eyes very large, centrals less than a

diameter ajiart, scarcely Vt, diameters IVom

til'' laterals; anterior portion of ocular area

almost vertical ruJicajiilJns Simon.

b-. Posterior eyes moderate in size, centrals over
a diameter apart, at least two diameters
from the laterals. Anterior jjoi'tion of

ocular area jjorrect luonaaloidi s (). I'.-t'ainbr.

bh. Tibia' i. and ii. with 4. — '1 long si)iues beneath.
«•'. Protarsi i. and ii. with 3 — 3 long spines

beneatli, besides lateral and apical s|)ines. . (dhiilKn V. l'.-('anil)r.

i''. Protarsi i. and ii. with 6— o long spines

beneath, besides lateral and ajiieal sjiines.

a'K External sclerites of vulva evenly rounded

on their outer nuirgin, sinuously concave

on their innei margin, their ai)ices

(anterior pcn-tion) sliarj), with a mimite

cusp on their outer margin prolalus O. P.-Cand)r.

b*. External sclerites of vulva emarginate on
their outer sides, evenly convex on their
inner margins, their ajiices forming a
broad hook, not very sharp at the point,
without anyminute cusp canaliviilahiit F. P.-Cambr.

168 ME. F. PICKAED-CAMBRIDGE ON NEW SPIDERS. [Feb. 17,

Senoculus albidus, sp. nov. (Plate XIY. fig. 14.)

T3rpe 2 > ill coU. Brit. Mus. Total length 14 mm.

In general appearance this species is similar to others of the
genus, the legs being clothed with a lateral fringe of curving
simple silky white hairs. The abdomen is also clothed at the
apex on each side with long pure- white hairs. The vulva consists
of two large oblong convex sclerites convergent forwards, and
somewhat concave at this point, see Plate XIY. fig. 14, where
also the vulva of *S', parallelus Simon is figured (fig. 13).

Hah. Brazil ; Rio Janeiro.

Note. — OiS. purpureiis (E. Simon) and 8. darwini (Holmberg)
I cannot give any reliable characters. S. inaronicus Tacz., the
type of the genus, is, according to the author, immatvire, and no
mention is made of the number of spines on the legs.

EXPLANATION OF THE PLATES.
Plate XIV.

Fig. 1. TJiaumasia velox, $ , vulva, p. 154.

2. „ annulipes, $ , vulva, p. 154.

3. Dossenus marginatus, $ , p. 155.

4. „ „ $ , vnlva.

5. „ „ ? , eyes.

6. Paradossenms nigricans, $ , p. 155.

7. „ „ ¥, vulva.

8. „ „ ? , eyes.

9. „ J) (? 3 tibial spur of palpus.

10. Thanatidius spinipes, ? , eyes, p. 156.

11. „ „ ditto, in front.

12. „ „ ? , vvilva.

13. Senoculus parallelus, $ , vulva, p. 166.

14. „ albidus, $ , vulva, p. 168.

Plate XV.

Fig. 1. TrecJialea Iceyserlingi, ^ , p- 163.

2. „ „ vulva.

3. „ MWMaior, ^5 tibial spur of palpus, p. 161.

4. „ „ ?, vulva.

5. „ „ $ , embolus of palpus.

5 a. „ „ (? , „ from below.

6. „ ellacomhei, ? , vulva, p. 161.

7. „ maccowMeZZi, (j, tibial spur of palpus, p. 162.

8. „ ,, (J, embolus of palpus.

9. „ connexa, $ , tibial spur of palpus, p. 162.

10. „ ); (? ) embolus of palpus.

11. „ ea-^emsa, <J , tibial spur of palpus, p. 162.

12. „ „ (J , embolus of palpus.

13. „ longitarsis, $ , embolus of palpus, p. 160.

14. „ „ (? , ditto, apex enlarged.

15. „ „ (J, tibial spur of palpus.

16. „ „ (J , ditto, example from Venezuela.

17. „ „ ? , vulva, example from Colombia.

18. „ amazonica, <? , tibia of palpus and spur, p. 163.

19. „ „ $ , embolus of palpus.

20. „ „ ? , vulva.

21. Hesydrus Jiahilis, $ , tibia of palpus and spur, p. 165.

22. „ jpalustris, $, tibia of palpus and spur, p. 165.

23. „ „ c? J embolus of palpus.

24. „ „ $ , riglit mandible, from outer side.

25. „ „ ?, vulva.

P Z.S. 1903, vol.1 PI .XVI.

Figs. 1,2. 3. 4. 7. 8. PHYLLOCPMCTOPTERUS ELIOTI
,, S.6.9 do. PICTUS

12 P.Z.S.1903.vol.I.P).XVJI

10,

■ Th.C

of ppcL -

TTh or pp

C.C del.

-v.od'

iSale &:DamelssoiiI.\'^ litK.

PfTYliOCIiyETOPTERUS ELIOTI

1903.] ON THE MARINE FAUXA OF ZANZIBAR. 169

2. On the Marine Fanna o£ Zanzibar and British East Africa,
from Collections made by Cyril Crossland in the Years
1901 and 1902.— Polych^ta. Part I. By Cyril
Crossland, B.A., B.Sc.^

[Received Jaimaiy 15, 1903.]
(Plates XYI. & XVII.^)
T/iirochcction.
The collections referred to were made under the following
circumstances : — Sir Charles Eliot, K.C.M.G., C.B., late Fellow
of All Souls College, Oxford, at present H.M. Consul-General at
Zanzibar and Commissioner for British East Africa, took me out
with him as his private assistant in his researches on Nudibranch
Mollusca. I made collections not only of this, bvit of the other
marine groups ; accounts of which, by various specialists, will
appear in these 'Proceedings' from time to time. The largest
collections are those of the Nudiln-anchiate Mollusca and Polychrete
Ann-elids, of which groups about 150 species in each are to be
described. Su^ Charles Eliot has ah^eady publi.shed one part of the
results of his examination of the Nudibi-anchs (P. Z. S. 1902, vol. ii.
p. 62) ; other papers by various authors wih shortly be ready.

For the benefit of other possible woi'kers in this region, I may
mention that the greater part of my collections was made in two
localities, viz. Chuaka Bay, on the east coast of Zanzibar, and
Wasin Harbour, near the Anglo- German boundary on the main-
land. ^ The former locality is extremely rich in shore forms, and
dredging in 3 fathoms of water at the north side of the mouth of
the bay was often very productive. Wasin Harbour averages a
depth of 10 fathoms, and here I collected almost entirely by
dredging. The bottom is extensively covered by a species of
Telesto, the branches of which are overgi'own by an encrusting red
sponge. Among this a great variety of the smaller Polycha?ta
and Nudibi-anchiata are found.

My mention of these two localities is of the more importance to
future workei's because large stretches of the coast are extremely
barren. Unfortunately this applies to the two principal towns of
East Africa, and especially to Mombasa. A few miles from
Zanzibar, e. g. round the islands and sandbanks which surround
the harbour, and on one portion of the shore about a mile to the
south of the town, at low spring-tide only, are lich collecting-
grounds, and near this latter area dredging is profitable, especially
at a depth of 5 fathoms. The greatest portion of the Zanzibar
channel is, however, extremely barren.

More detailed descriptions of the reefs of East Africa, with
maps, will be found in my two papers : " On the Coral Reefs of
Zanzibar," and " Pemba and British East Africa," in the Proc.
Camb. Phil. Soc. vols. xi. & xii. (1902).

1 Conimnnicated by Prof. W. C. JIcIntosh, F.Z.S.

2 For expLinatiou of the Plates, sec p. 176.

170 MR. CYRIL CROSSLAND ON THE [Feb. 17,

The gratitude of all zoologists is thus due to Sii^ Chai'les Eliot,
whose generosity and scientific zeal have enabled these collections,
the first of considerable size from this region, to be made. I
wish also to exjoress my best thanks to Mr. Stanley Gardiner, who
helped me by his advice and in every other possible way, both
during my residence in East Africa and in the working out of the
collections on my return to England.

Ch^topterid.e.

Genus PnTLLOCHiETOPTERUS.

The position of the genus Phyllochcetoptents among the lower
Chpetopteridfe is shown by the following table of the genera : — •

A. Notopodia of the second body-region not foliaceous

{i. e. two body-regions onl^') Ranzania.

B. With foliaceous notopodia posteriorly.

1. Body divided into two regions Telepsatus.

2. Body divided into three regions.

C One pair of tentacles Spioch^topterus.

<.Two pairs of tentacles Phtlloch^itoptekus.

The most characteristic feature of the Chsetopteridse is the
adaptation of certain parapodia for the production of a respiratory
current, which modification, completed in the genus Chcetopterus,
makes the latter one of the most remarkable forms of animal life.

This peculiarity is not developed in Ranzania, whose only
Chsetopterid features are its general build of body, which is
divided into an anterior flattened muscular and glandular portion
bearing long notopodial setse only, and a rounder weaker posterior
portion with delicate notopodia and neuropodia of uncinigeral
tori ; in addition it is tubicolous in its mode of life, and procures
food by the ciliated grooves of its tentacles and dorsal surface.

To these features are added notopodial gills in Telepsavus, in
which case they are developed on every segment of the hind-body.
In the remaining two genera this modification is restricted to
more or fewer of the middle segments, to two only in Spiochceto-
pterus and several species of PIiTjllochcetopterus, but up to 25 in
other species of the latter.

In all the eight species of PJiylloclicetopterus yet known the
body is vermiform, the notopodial gills comparatively small and
simple, bifid, and containing capillary setse. All the setpe are
characteristically Chsetopterid, in the first region long with leaf-
like ends, and arranged in a row as in Chcetopterus, though the
notopodia of P. acictoUgerios alone approach in their long pointed
shape those of the former genus. In the fourth notopodium of
the first body-region one or more setee are enlarged and of a dark
brown colour, but are not flattened as in Chcetopterits. The uncini
are of one kind only throiighout the body, but their form is more
specialized than in Chcetopterus, and is that which occurs in, for
example, the Sabellidse. The parapodia ax-e alike in all the sj)ecies,
differing only in the proportions of the parts, excepting the hind-
body notopodia of P. aciculigerus (6).

1903.] MARINE FAUXA OF ZANZIBAR. 171

The prostomium is, in this family, usually more or less reduced.
In Chcetopteriis itself it is completely merged with the oral
funnel, and the same condition is described for some Phyllocha-to-
pterids (e. g. P. claparedi IVIcIntosh). In all the members of this
genus it is normally scarcely distinguishable externally, both pairs
of tentacles being carried by the peiistomium. In certain specimens
of F. elioii, however, as described below, it is quite prominent,
and so was discoverable in all the four species I liave examined.

Tlie first body-region is an almost solid mass of muscular and
glandular tissue in all species, but jiosteriorly musculature is
confined to the ventral surface, except at the level of the parapodia,
where muscle-bands surround the body. Between these the body-
wall is excessively thin doi-sally and laterally, and always black
fi'om the colour of the gut-walls.

Up to the present six species have been described, viz. : — ■

P. socialis Clapar^de. Mediterranean and Atlantic (1) (4).

P. majoi" Clapar^de. Mediteiranetin (1).

P. gracilis Langerhans. Canary Islands (2).

P. claparedi Mcintosh. Japan (3).

P. aciculigems Crossland. The IMaldive Islands (6).

P. gardineri Crossland. The Maldive Islands (6).

To these I now add : —
P. elioti. Zanzibar.
P. piictus. British East Africa.

Clapar^de's P. fallax was founded on such chai'actei'S as the
numbei'S of segments composing the three body-i-egions, the
i-inging of the tube, &c. ; these have been shown by Roule (4) to
be variable in members of this genus.

The principal distinctions of systematic importance are : —

(1) Presence or absence of eye-spots, and of the development

of the prostomium,

(2) The number of the strong setfe in the notopodium of the

fourth setigerous segment.
(.3) The proportionate sizes of the paiis of the parapodia of

the regions B and C
(4) Ringing and other such details of the tubes are variai)le,

but (1) in some species they are sti-aight and occiu"

singly, and (2) in others they are of a U -shape and occur

in numbers twisted together.

These characters are distributed as follows : —

(1) Eye-spots present in P. socialis, P. major, P. elioti, P.

gardineri, P.pictus; absent in P. claparedi, P. acicidigerus,
P. gracilis.

(2) A single strong seta in 4th notopodiiun in P. socialis \

P. major, P. claparedi, P. pietus. More than one seta
strengthened. P. gracilis with three, P. gardineri three,
P. elioti two, and P. acicidigerus eight.

1 In P. sncialis two setie may exceptionally occur, hut the extra one is alwaj's
smaller than the one it accompanies.

172 MR. CYRIL CROSSLAND ON THE [Feb. 17,

(3) The parapodia of the region consist of double neuro-

podial tori and a clavate papilliform notopodium containing
a single seta in all the species except P. aciculigerus and
P. claparedi.

(4) ^ Tubes straight, occurring singly in P. claparedi, P. elioti,

P. onajor. Twisted together, in numbers, P. socialis and
P. pictus.

Phylloch/etopterus elioti", sp. n. (Plate XYI. figs. 1-3 & 8 ;
Plate XVII. figs. 10-13.)

Of this species numerous specimens occur in Chuaka Bay,
Zanzibar. The sti-aight, stiff, opaque black tube, 6 to 9 inches
long, is buried in the sand at low tide, one or two inches only
projecting. Its presence is usually rendered conspicuous by the
growth of a tuft of bright green enteromorpha on the projecting
portion.

The colour of the living animal is milk-white anteriorl}^, and
black posteriorly. There is no pigmentation, and the brilliant
blue and yellow colours described by Clapai'ede and Lo Bianco in
the Naples species are not found in either of the two species I
have seen alive. The black colour of the gut gives the vTSiial
green solution in alcohol.

The peristomium is very mobile, the whole shape of the head
being thus very different in difierent specimens. When expanded,
its shape is as in PI. XYI. fig. 1 . In a few cases only, where the
head is much contracted and bent back, does the prostomium and
its eye-spots come into view as in fig. 2.

The first pair of tentacles are very slender and long, attaining
a length of 9 mm.

The fore-body measures, in a large specimen, 6 mm. in length
by 2'5 mm. in breadth, and consists of about 14 setigerous
segments. (There were 13 in three specimens, 14 in three more,
and 15, 16, and 17 each in three more. The average is thus 14
in nine specimens.) The notopodia are very short and stumpy,
but the setfe and their arrangement, as in the other species of the
genus, recall Chcetopterus. The fourth foot has two strong setfe
on each side, rarely three on one side or the other, though in one
specimen there were three on one side and four on the other.
The ordinary setse are all straight.

The region B (fig. 3) consists of from 20 to 25 segments, the
parapodia of which are very like those of P. gardineri, except that
the notopodial flap is slightly smaller, and there is a long space
between it and the ventral portion of the neuropodium, and the
parts of this latter are of more nearly equal size. The neuropodium
of the first segment of the gill-region is not divided. There are in
each notopodium five long setse whose flexible heads project. The
uncini (fig. 8, c) are triangular with very fine teeth, just visible
under a ^-inch objective.

The hind-body has more than 25 segments, but none of my

1 The tubes of Gardiner's two species were uiifortuiiatelj^ not obtained.

2 Thus named in honour of Sir Charles Eliot, K.C.M.G., C.B., H.M. Consul-
General at Zanzibar, and Commissioner for British East Africa.

1903.] MARINE FAUXA OF ZAXZIBAR. 173

specimens are quite complete. The notopodia are the usual
clavate papilla? containing a single seta. The neuropodia ai-e as
in the gill-be<aring region, but much smaller.

The two species P. elioti and P. ga7-dineri are closely related,
differing only in the reduction of the peristomial collar in the
latter and the proportions of the mid-body parapodia and sette.

The only work on the intei-nal anatomy of a member of this
genus is the section of the fore-body of P. claparedi given by
Mcintosh, although examination of tlie anatomy of tlie species
of Pliylloclufiiopterus would be extremely interesting, both for
comparison among themselves and with the other geneiu of the
family \ especially Chcetoj)terus, Having neither time nor full
opportunity for this woi-k, to which my inclinations ai-e most
strongly drawn, I can only give hei'e a few deductions from the
examination of a series of sections of a specimen of this genus,
which appear to differ in many minor details from the above-
mentioned section of P. claparedi.

Body-ioall. — The region A is, as shown by the sections figured
(PI. XVII. figs. 11 & 12), an almost solid mass of glandular
and muscular tissue, in marked conti'ast to the delicacy of the
i-egions B and C. The whole of the ventral epithelium is sti'ongly
glandular beyond the limits of the well-mai-ked ventral shield, and
these glands extend anterioi'ly to the dorsal surface and even on to
the prostomium. The cuticle is of extreme delicacy, if really
present. The musculature consists of weak circular and very
strong longitudinal muscles, but the division of the latter into
two dorsal and two ventral bundles cannot be made out. In the
regions B and C the only muscle occurinng doi'sally is a very
delicate circular layer (PI. XVII. fig. 13). Diagonal fibres can be
ti'aced in the region A between the nei've-cords and the notopodia,
which become more definite muscles in the region B (fig. 13).

The nervous system is in contact with the skin. The two
ventral cords lie at a distance from one another in the region A,
but approach one another posteriorly (cf. figs. 12 & 13), their
arrangement corresponding thus to that in Cluciopterus. Trans-
verse commissures, portions of one of which are shown in PI. XVII.
fig. 12, connect the two cords.

The brain (PI. XVII. fig. 10), which in Chcetopterus is a narrow
band dorsal to the mouth, not differentiated from the circum-
cesophageal commissures, is here more distinct, in correspondence
with the presence of the prostomium in this genus and its absence
in the case of Choitojiterus. It is a perfectly simple swelling of the
circum-oeso]ihageal commissures. The continuity of the nervous
system with the epidermis is shown by a comparison of the
sections of the prostomium given in figs. 10 & 11. The former
shows the circumoeso2)hageal connectives, that on the left as it
crosses from the venti'al to the dorsal side, that on the right, near
the eye-spot (e) passing into the prostomium. The z-eplacement
of the ordinary glandular epithelium of this area by the densely

* For anatomy of Telepsavus, sec pi. xiii. and its explanation in Claparede's
' Annelides ISedeutaires.'

174 MR. CYRIL CROSSLAND ON THE [Feb. 17,

staining nerve- nuclei of the brain (which are similar to those
seen on the outer side of the ventral nerve-cords &c.) is shown
distinctly, this replacement being complete in fig. 10, where the
epidermis is composed entirely of nerve- cells.

The eye- spots are a pair of groups of cells, each containing
numerous minute granules of black- brown pigment. These are in
continuity with the nerve-cells of the biuin, but are anterior and
dorsal to its fibrous part.

The coelom is small and broken up into several distinct portions.
The largest of these are a pair of spaces lying laterally and
ventrally to the gut (fig. 12), bounded dorsally by the powerful
muscles of the seta-sac. Dorsally and medianly, commencing as a
space in the prostomium, is a third pai't which posteriorly becomes
a mere canal surrounding the doi'sal blood-vessel. In the region B
the coelom is more normal, thoiigh reduced in size. It remains
divided into right and left halves by the dorsal and ventral
mesenteries of the gut.

The vascular system consists of dorsal and ventral vessels, both
running in the gut-mesenteries. The former breaks up at the
base of the prostomium into three branches (fig. 10), and large
connecting-vessels are found in the anterior segments. Posteriorly
the dorsal vessel forms a laig-e sinus covering the dorsal wall of
the gut (fig. 13).

Alimentary canal. — The mouth is richly ciliated, and the outer
ends of its columnar epithelial cells contain a few minute specks
of the black-green pigment so characteristic of the family. These
are absent from the narrow triangular gut of the region A, but
reappear in great quantities in the larger thick- walled alimentary
canal of the regions B and C (fig. 13).

The gut-lining here consists of long crowded cells, the swollen
distal ends of which are ci-owded with minute specks of the
pigment, indicated by dark dots in the figui'es.

The transverse sections of the notopodial gills, shown in fig. 13,
are interesting. The space which they contain is ccelomic, and at
its centre is a'bundle of five or six setae wrapped closely together
by fleshy tissue. The transverse section of this resembles that of
a telegraph-cable, the wires being represented by the set*. At
each end of the oval section is seen a sharply-cut groove, the
sides of which are of granular, rather deeply- staining protoplasm,
without nuclei, supporting externally very long cilia. Laterally
is a more deeply-stained area of the epithelium, perhaps of nervous
function.

Phylloch^topterus picttjs, sp. n. (Plate XYI. fig. 5.)
The tubes of this species are brown and translucent, quite free
from sand or mud. They were found clustered together in con-
siderable numbers on the under side of a large stone at low-water
level on Pungutiayu islet, Wasin. Each tube has two openings,
each limb of the U being about 4*5 cm. long. The bending is
irregular, and they are twisted and fused together, so that it is
usually almost impossible to separate out any one tvibe.

1903.] MARINE FAUXA OF ZAXZTEAR. 175

The contained worms are the smallest of the species descrilied
here, or in (6), being slightly smaller everyway than in P. elioti.

The specific name refei'S to the abundance of bi-own colour
found on their tentacles and fore-body. Pigmentation, except of
the gut, is rare in the ChK>topterida\ Of the other species, it is
descril)ed in only two, viz. F. clcqxiredi (3) and P. socicdis (1) : in
both these it is comparatively slightly developed.

The ground-colour of tlie anterior region, tentacles, and para-
podia of P. jnchis is, in life, creamy white. The greater part of
the posterioi- regions is, as usual, black, by reason' of the pigmen-
tation of the gut.

On either side of the groove which runs along tlie upper surface
of tlie larger tentacles are regularly arranged squarish blotches
of brown, as shown in PI. XVI. fig. 5. Brown dots are scattered
also over the pro- aiid peristomium and the anterior segments of
the fore-body. There are two dark bands with indefinite edo-es
along the ventral bases of the notopodia, and a broad crescent
across the ventral surface of the end of the region A.

The movith is not at all funnel-shaped, but rather slit-like, tlie
two small peristomial lappets coming together from either side.
The prostomium is large, flattened from side to side, and projects
above the dorsal surface of the fore-body. The very slender second
setigerous pair of tentacles lie close on either side of it, being thus
very inconspicuous. Two elongated, but distinct, eye-spot?, of a
dark brown colour, occur one on each side of the prostomium.

The legions A and B consist of the following numbers of
segments in difl^erent individuals : —

A ... 15 16 — 13 15 12 15 1*3

B... 7 9 5 7 5 3 8 —

* These specimens were below the average size.

The fourth notopodium has but one large seta of the shape
shown in PI. XVI. fig. 9, a. The bending of this seta is remarkable
as occurrnig only in this species. Its three teeth, when seen from
one point of view, giwe an explanation of Langerhans's figure of
the corresponding seta in P. gracilis, which is not otherwise
intelligiV)le.

The other setae are all straight, and present no peculiarities
except those of the last notopodium, which are bent, and the head
is finely toothed (fig. 9, c).

The parapodia of the region B (figs. 5 & 6) are small, the
neuropodial ridges especially so. In correspondence with this the
uncini are very minute and delicate (-03 mm. long), and their
teeth barely visible under a |-inch objective (fig. 9, d). The
neuropodia do not extend far up the side of the body, and the
space between them and the notopodial gills is not filled up by
the triangular membranous gill-flap as in the other species (fig. 6).
The notopodia contain two or three thin seta? which do not project."

In the region the notopodia are reduced to little clavate
papilla?, containing one seta, as in all the other species except

176

ON THE MARINE FAUNA OF ZANZIBAR,

[Feb. 17,

P. daparedi and P. acicidigerus (see fig. 6, of P. elioti). The
neuropodia also are small.

List of the Literature.

(1) Claparede. — Annelides du Golfe de Naples, 1868.

(2) Langerhans. — " Ueber einige Canarischen Anneliden." Nov.

Act. der Leop.-Carol. Deutschen Akad. Bd. xliii., 1881.

(3) McIntosh.—' Challenger ' Reports, vol. xii., 1885, p. 374.

(4) Roule. — " Oampagne du Caudan." Ann. d. I'Universite de

Lyon, 1896.

(5) Lo Bianco. — Ann. Tub. nel Golfo di Napoli, 1893.

(6) Crossland. — J. Stanley Gardiner's ' Fauna and Geography

of the Maldive Archipelago.' In the press.

EXPLANATION OF THE PLATES.
PlATJS XVI.

Fig. 1. Normal condition of the head of Plii/llochcetopterus elioti (p. 172), as seen
from dorsal side. X 10.

2. Side view of a specimen whose prostomium and eye-spots are visible. X 10.

3. Side view of three segments of the region B. X 10.

4. Ventral view of the same.

5. P. pictus (p. 174). Dorsal view of anterior part of hodj\ X 10.

6. Side view of three segments of the region B. X 10.

7. A notopodium of the region C in either species. X 100.

8. Setffi of P. elioti (p. 172).

(a) Thickened seta of fourth parapodium. X 83.
(6) Normal seta from third foot. X 90.
(c) An uncinus. X 1000.

9. Seta3 of P. picttis (p. 172).

(a) Thickened seta from the fourth foot. X 100.

(h) Head of a normal seta from the same bunch. X 200.

(c) From last foot of region A. The shaft is bent and part of the head

finely toothed. X 100.
{d) One of the uncini. X 1000.

Plate XVII.

Fig. 10. Section of Fhyllochcetopterus elioti (p. 172) through the brain, bases of
short tentacles, &c.

11. Section a little anterior to fig. 10, through mouth, first feet, &c.

12. Section at level of fifth foot.

13. Section near beginning of region B. All sections X 42.

The meaning of the lettering on the figures of the Plates is as follows :—

6.^).=blood-vessel.
cJrc.TO.=circular muscle.
COS. =ccBlum.
(^.6».= dorsal blood-vessel,
(^.ct?.= dorsal ciliated groove.
c^.cc6.=dorsal division of ccelom.
«:Lwettr.= dorsal division of the neuro-
podia.
c.=eye.
5'.=gut,
^iZL=gill-flap between noto- and
neuropodia.
^L= glandular areas or epithelia

&c.
Im. = longitudinal muscle-layer.
5w.= muscle.
iMO.=mouth.
m. qfppd.=mMscles of parapodium.

«•.= nerve.
w.c.=nerve-cells.
n.cd. = nerve-cord.
»»/■.= nerve-fibres,
wejo/j. =nephridium.
weMr.=neuropodium or ventral division

of the foot.
wo^o.= notopodium or dorsal division

of the foot,
joeri. =peristomium.
^ji3f^.= parapodium or foot,
jsro. =prostomium.
s.=seta.
ti &io = first and second pairs of ten-
tacles.
■y.6«.= ventral blood-vessel.
•i).MeM>',= ventral division of the neuro-
podia.

p. Z . S. 1903, ATolIPLXVIII.

B..3rooinael.
M.P.T>a.rker ]ifk.

O.C.

cd>...

^

Zdt.

--- h-dt:

7^.at•

- cck. o.p.-

V-o.p-

Pourkeir & "West inap.

AXIS, ATLAS, & PROATJLAS OF
GOMPHOGNATHUS k TRIRACflOD ON.

1903.] ON THE Vi:UTEnil.K OF TIIERIODOXTS. 177

3. Oil the Axis, Athis, ami Froatliis in the Higher
Theriodonts. By R. Bkoom, M.D., B.Sc, C.M.Z.S.

[Rocoived Jaiuuiry 20, 1903]

(Plate XVIII.")

Hitherto, tliouyli something has been known of tlie structure
of the axis in at least two of the Dicynodont.s and in (J>/iiO(jiiathus,
we have been practically ignorant of the structure of tlu; atlas in
any of the Theriodonts or Anoniodonts. In fact, Prof. iSeeley (1)
in descrilnng the axis of CyaoynatJius, seems to be in doubt
whethei- he is really dealing with an axis, or with an axis and
atlas cond)ined. On page 100 of his paper he says, "The first
vertebra appears to be anch}dosed to tlie second " ; while a little
further on he states that " this vertebi'a is remarkable for its
form being exactly like the odontoid pi'ocess of the vertebra in
many animals, and suggesting the idea that the atlas is lost " ;
aud in a note he adds : '' This appears to be confii-med by the sub-
setjuently to be described condition in Tropidostomci duiini." In
counting the vertebrse, however, he counts the axis as the 1st
cervical, and states that "there are 6 cervicals." Gadow (2), in
his recently published work on Reptiles, possibly misled by Seeley,
definitely states that " the atlas is fused Avith the axis." Con-
sidering how mammaldike the higher Theriodonts are, and that
the axis bears a very marked resemblance to the mammalian axis,
having a large spine and a well-developed odontoid process, it
seems remarkable that the idea should have arisen that the atlas
was anchylosed to the axis, more especially as there is no part
of the axis in Cijnognathios that bears any resemblance to any
known atlas.

When in Grahamstown recently, I had an oppoitunity, through
the kindness of Dr. Schonland, of examini)ig the Theriodonts in
the Albany Museum, and especially the very fine specimens of
Gomjjhognathus and Trirachoclon^ which have been described by
Seeley (3), and which have been so marvellously developed under
his direction. In both the type specimens of Gomphognathus
kannemeyeri and Trirachodon kannemeyeri the anterioi' cervical
vertebrae are most beautifully shown, but Seeley in his desciiption
of the specimens makes no i-eference to this most important
region, beyond stating (p. 53) that " the occipital plate [in Trira-
cliodoii\ is not completely exposed owing to some of the vertebrae
being in contact." Two of his figures of the skull of Trirachodon
show the atlas and part of the axis ; but as no distinction is shown
between inati-ix and bone, it is ilifiicult to make anything of the
drawing.

The beautiful specimen which forms the type of (Joinphogiudhns

' For explanation of the Plate, sco ji. 180.

Proc. Zool. Sue— 1S)03, Vol. I. Xo. Xll. 12

178 DR. E. BROOM ON THE [Feb. 17,

consists of two complete mandibles in articulation with the
posterior half of the skull. The lower jaw lies almost at right
angles to the axis of the skull, and the upper cervical vertebrae
have been to some extent protected by passing down between the
two rami. The skull consists of the practically complete occiput
with both tempoial arches in almost perfect condition, and with
most of the parietal and sphenoidal regions well shown. Though
the sutures unfortunately are not distinct, the occiput is so well
preserved, practically without distortion, that I think it worthy
of being figured.

The vertebrae so far as preserved consist of the proatlas, atlas,
axis, and the 3rd and 4th cervical vei'tebrae.

The atlas {at.) consists of a well-developed arch and a distinct
hypapophysis {h.at.). The arch is partly broken in the middle, but
must have borne considerable resemblance to that of the mammals.
It differs in having had a distinct posterior zygapophysis {z.at.)
for articulation with the small anterior zygapophysis of the axis
{z.ax.). The zygapophysis of the left side is well presei-ved, but
that of the right has been broken off. On the right side the
outer part of the arch is continued backwards as a well-developed
bony process or rib {r.at.). It is probable that the process is
really a rib, but I have been unable to find a suture between it
and the arch. On the under surface, what appears to be part of
the rib of the left side is seen lying by the side of the body of the
axis {r.at.). The well-developed hypapophysis is also well seen.
It appears to have had only a ligamentous connection with the
arch of the atlas.

In front of the arch of the atlas, and lying in the hollows formed
above and outside of the occipital condyles, are the two portions
of the proatlas. Each consists of a short bony arch with a
well-developed j)rocess passing upward and outwai'd fi'om its
outer end. ^

The axis is well developed, and bears considerable resemblance to
that of Cynognathus. The spine forms a very large flattened crest.
In the specimen it is somewhat damaged, but there is probably
not a great deal missing. In front, the spine passes forwards
over the arch of the atlas, and possibly forms an articulation
with it as in the Monotremes. Between the anterior part of the
spine and the odontoid process is the small anteiior zygapophysis
{z.ax.). The odontoid process (o.j».) is well developed, and on the
right side it is seen articulating with the arch of the atlas. On
the under surface the body of the axis is seen, as also a part of the
odontoid process in articulation with the hypophysis of the atlas.
On the upper surface, near the posterior end of the centrum, on
the right side is a portion of the rib of the axis.

The 3rd and 4th cervicals present no features worthy of any
special remark. Remains of ribs are seen in connection with
each.

In the type specimen of Trirachodon Jcannemeyeri there are
preserved and well shown the two portions of the proatlas, the

1903.] VERTEBR.E OF TIIERIODOXTS. 179

arch and liypapophysis of the atlas, and the odontoid process and
part of the body of the axis. Tliough the structure of the
vei'tebrse is essentially similar to that in Gomphogiiathus there are
many points of diSerence.

The proatlas {2).a.), though occupying the same relative position
as in Gomphoy)iat}tas, is less specialized. It is present as a pair of
curv^ed, moderately thin, bony plates lying in fi'ont of, and i^robably
somewhat overlapping, the ai-ch of the atlas. There is no bony
process developed as in Uomi)liognatlius.

The atlas («^) consists of a well-developed arch and a distinct
hypapophysis. The arch is not so complete as in Gompliof] iiaihas,
anil must have differed considerably in regai'd to the airangement
of the antei'ior articular surfaces. In Gompliognathiis the occi-
pital condyles {p.c!) are small and moderately close together,
whereas in Trirachodon the condyles {p.c)) are veiy lai-ge and as
wide as the atlas. The arch, perhaps on this account, does not
close in inferiorly to meet the hypapophysis, with which it can
only have been attached by ligament. Tiie zygapophysis {z.at^
for the axis is very small. The hypajjophysis {Ji.at.) is very
similar to that in Gomphognatlius.

Of the axis only the odontoid process, imperfectly displayed,
and a part of the body remain.

The structure of the upper cervical vertebrae is so imperfectly
known in the majority of fossil I'eptiles, that there is some diffi-
culty in satisfactorily dealing with the affinities of the structures.
It is difficult to avoid being struck by the close resemblance of
the axis, atlas, and proatlas in the Theiiodont to those structures
in the Crocodiles, more especially as there is no close affinity
between the groups. The explanation is probably to be found in
the fact that both the Crocodiles and tlie Theriodonts, though far
removed from each other, have retained with but slight modifi-
cation the type of structures met with in their common ancestor.
In Hphenodon, though the proatlas is small, the tj^pe is practically
the same ; and I have recently discovered that Procolophon has a
well-developed proatlas. Frocolophon has usually hitherto been
associated with the Dicynodonts, Theriodonts, and Pareiasaurians,
but as it has a persistent notochord, abdominal ribs, a plate-like
pubis and ischium, and 2, 3, 4, 5, and 4 phalanges in the digits, it
is manifestly much more closely lelated to PalcnohatUrla than it is
to the Tlieriodonts and Dicynodonts. In most of its structures it is
probably as primitive as the common ancestor of the Crocodiles
and Theriodonts, so that there is good reason for believing that
the common ancestor had a well- developed proatlas and pi-obably
an atlas and axis veiy similar to those found in later forms,
but almost certainly with the odontoid process as a distinct
element, as seen in Ichthyosaurus. Ko proatlas has yet been
detected in Pareiasaurus, but I have found one in Lystrosaurus
{^ = Ptychognaihas Owen). It will probably yet be found in mo«t
of the primitive reptilian types.

12*

180 MR. C. TATE REGAN ON THE [Feb. 17,

Wo7'ks referredj to.

(1) H. G. Seeley : " On the Skeleton in new Cynodontia from

the Karroo Rocks." Phil. Trans, vol. 186 B. p. 59 (1896).

(2) H. Gadow. ' Amphibia and Reptiles.' London, 1901.

(3) H. G. Seeley : '' On the Gomphodontia." Phil. Trans, vol. 186 B.

p. 1 (1896).

EXPLANATION OF PLATE XVIIL

Fig. 1. Occiput and upper cervical vertebras of G-ompliognathus Tcannemeyeri. Nat.

size.
Fig. 2. Under view of atlas and axis of G-ompliognatlnts. Nat. size.
Pig. 3. Posterior view of atlas &c. of Trirachoclon Icannemei/eri. Nat. size.
Fig. 4. Under „ „ „ „ ,, Nat. size.

Fig. 5. Side „ „ „ „ „ Nat. size.

at., atlas ; ax., axis ; e.o., exoccipital ; li.at., lij'papophysis of atlas ; o.c, occipital
condyle; o.p., odontoid process ; jp.a., proatlas ; r**, r'*, ribs ; r.at., atlas rib; r.ax.,
axis rib; sq., squamosal; z.at., zygapox^liysis of atlas; z.ax., zygapoplij'sis of axis;
3c, 4o, 3rd and 4th cervical vertebrae.

4. A Eevision of the Fishes of the Genus Triacanthus.
By C. Tate Regan, B.A.^

[Received January 20, 1903.]

Although six species of Triacanthus were described and figured
by Bleeker in the ' Atlas Ichthyologique,' GUnther, in his Catalogue,
recognized only three (viz. : Tr. strigilifer Cantor, biacideatus
Bloch, and hrevirostris Schlegel), and did not accejat either of the
four described by Dr. Bleeker as new, but placed three of them
(viz. : Tr. macriirus, hlochi., and oxycephalus) in the synonymy of
Tr. hiaculeatus^ and the fourth (^Tr. nieuhofi) in that of Tr. hrevi-
rostris. Since then this arrangement has not been challenged, nor
has any new species of this genus been described.

Subsequent to the reading of my paper on the Plectognathi ^, I
examined the specimens of Triacanthi in the British Museiun
Collection, which include Bleeker's types, and I have come to
the conclusion that all six species described by Bleeker are valid,
although the one he called Tr. niacnvrus is certainly identical with
Tr. hiaculeatus Bl., a species not recognized by him ; whilst a
seventh species, which has been generally confused with Tr. hrevi-
rostris Schleg., is now described for the first time as Tr. indicus.
A complete revision of the synonymy has thus become necessary,
and the need for more complete diagnoses of the various species is
obvious. In the descriptions given below, which are in each case
based on several specimens, the total length is measured to the
base of the first caudal ray, the length of head to the gill- opening,
the length of the caudal peduncle from the base of the last dorsal
to that of the first caudal ray, the length of the snout from its tip
to the vertical from the anterior margin of the eye, that of the

1 Communicated by G. A. Boulengee, F.R.S., V.P.Z.S.
" P. Z. S. 1902, ii. p. 284.

1903.]

FISHES OF THE GEXUS TIIIACAXTIIUS. 181

postor1)ital part of the head from the vei-ticnl from tlie posterior
margin of the eye to the gill-opening. Young specimens have a
shorter snout, a deeper body, and longer dorsal and ventral spines
than adults, and have been excluded fi-om the diagnoses. In each
case the total length of the largest example here described is given.

Triacaxthus Cuvier.

Body compressed, covei'ed with small rough scales ; caudal
peduncle more or less elongate, tapering ; mouth small ; teeth in
2 sei'ies in each jaw, 10 in each outer sei'ies, incisor-like ; those of
the inner series obtuse, rounded, 6 in the upper j:iw, 2 in the
lower. Branchiostegals 6 ; pseudobranchite present. Spinous
dorsal with 5 rays, the first a long and strong spine ; soft dorsal
with 20-25 rays; anal with 15-20 rays; ventral fins each repre-
seiited by a strong spine, without soft rays ; caudal widely forked.
Vertebrae 20.

In all the species the colour is strikingly similar, being bluish
grey above, silver)^ below. The only tangible difierences in colour
are that the membrane of the spinous dorsal fin is in some specimens
immaculate, in otheis wholly or pai-tly blackish.

Key to the Species.

I. Second raj' of spinous dorsal more than half as

long as tlie first, membrane of spinous dorsal
immaculate ; length of base of anal half that of
soft dorsal; 1). V, 20-22 ; A. 15-16 1. T>: sln<jilife)-CM\ior^

II. Second ray of spinous dorsal less than half as

long as the first.

A. Snout concave ; pelvis between ventral spines

tapering to a distinct point posteriorly.
a. jMcnibranc of spinous dorsal immaculate;

length of caudal peduncle 4}r-5| times iu

total length ; length of base of anal l^-lf

times in that of" soft dorsal ; D.V, 22-23;

A. 16-17 2. Tr. hJoahi Blocker.

h. Membrane between first two rays of spinous

dorsal blackish ; length of caudal peduncle

4.f-5 times in total length; length of liase

of anal Ij-l^ times in that of soft dorsal;

D.V, 23-24; A. 18-19 3- Tr. biaculcatus Bloch.

c. Membrane of spinous dorsal immaculate;

length of caudal peduncle 63-7 times in

total length; length of base of aiml Ij!-

IJ times in that of soft dorsal; D. Y,

21-25; A. 17-19 4. Tr. o.vi/cepJialus TAcckcr..

B. Snout straight ; pelvis between ventral spines

sairceU" narrowed posteriorly; D.V, 24-25;
A. 18-20.
a. Membrane of spinous dorsal blackish ; snout

half as long as the head ; postorbital part of

head not shorter than cye-diametf^r 5. T>: breviroslris ^chlogel.

h. Membrane of spinous dorsal blackish ; snout

more than halt as long as the head ; post-
orbital part of head shorter than eye-diameter. 6. Ti: iHiiifKS, n. sp.
c. Membrane between first two rays of spinous

dorsal blackish ; snout less than half as long

as the head ; postor6ital part of head shorter

than eye-diameter 7. Tr. nieuhofi Sleeker.

182 MR. C. TATE REGAN ON THE [Feb. 17,

Triacanthus STRIGILIPER.

Triacanthus strigilifer Cantor, Mai. Fish. p. 363, pi. ix. (1847);
Bleeker, Atlas Ichtli. v. p. 89, pi. ccxxix. fig. 3 (1865); Giintliei",
Cat. viii. p. 211 (1870).

Triacantlmis longirostris Hollard, Ann. Sci. Nat. (4) i. 1854,
p. 46, pi. ii. fig. 3.

Depth of body about equal to the length of head, abovit 3
times in the total length, length of caudal peduncle 4|-5 times.
Snout very slightly concave, its length 1|—1| times in that of the
head, eye-diameter 3-4 times, interorbital width 4-5 times.
Interorbital space concave, without distinct median ridge. Upper
edge of occipital ci'est in the same straight line as that of the
snout ; distance to base of first dorsal spine from posterior margin
of eye about 1^ times the eye-diameter. D. V, 20-22 ; A. 15-16 ;
the first dorsal spine longer than the head, the second considerably
more than llalf as long as the first, the others much shorter ; the
base of the anal fin about half as long as that of the soft dorsal
fin ; pelvis between the ventral spines moderately broad, tapering
to a point posteriorly. Membrane of spinous doi'sal immaculate.

Total length 185 mm.

Hah. (Seas of Arabia and India ; East Indian Archipelago.

Triacanthus blochi.

Triacanthus hlochi Bleeker, Nat. Tijds. Ned. Ind. iii. 1852,
p. 81 ; Atlas Ichth. v. p. 89, pi. ccxvii. fig. 1 (1865).

Triacanthus hiaculeatus (part.) Giinther, Cat. viii. p. 210
(1870).

Depth of body about 3 times in the total length, length of head
3|-3|- times, length of caudal peduncle 4^-5^ times. Snout
concave, its length 1|— 2 times in the length of head, eye-diameter
3-3| times, interoi'bital width 4-5 times. Interorbital space
concave, with median ridge scarcely, if at all, distinct. Upper
edge of occipital crest forming an angle of about 160° with that
of the snout; distance from posterior margin of eye to base of
first doi'sal spine about 1| times as long as the eye- diameter.
D. Y, 22-23 ; A. 16-17; the first dorsal spine longer than the
head, the others short; length of base of anal fin 1|-11 times in
that of the base of soft dorsal fin. Pelvis between ventral spines
narrow, tapeiing to a slender point. Membixme of spinous
dorsal immaculate.

Total length 150 mm.

llah. East Indian Ai'chipelago ;_ China.

Triacanthus eiaculeatus.

Balistes hiaculeatus Bloch, Ausl. Fische, pi. 148. fig. 2 (1785).
Triacanthus hiaculeatus Cuv. Begne An. ii. p. 152 (1817).
Triacanthus angustifrons HoUard, Ann. Sci, Nat. (4) i. 1854,
p. 45, pi. ii. fig. 2.

1903.] FISHES OF THE GEXUS TRIACAXTUUS. 183

Triacanthus macrwus Bleeker, Atlas Ichth. v. p. 91, pi. cexxii.
fig. 3 (1865).

Triacanthus hiaculeatus (part.) Giinther, Oat. viii. p. 210 (1870).

Depth of body about 24 times in the total length, length of
head about 3| times, length of caudal peduncle 4|-5 times.
Snout concave, its length 1^14 times in the length of head, eye-
diameter 3jT-4 times, intei'oi-bital width 4-5 times. Intei'oi'lntal
space concave, with median lidge scarcely, if at all, distinct.
Upper margin of occipital crest forming an angle of about 170°
with that of the snout ; distance fi'om postei-ior mai-gin of eye to
base of first dorsal spine lf-l|- times as long as the eye-diameter.
D. V, 23-24; A. 18-19; the first dorsal spine longer than the
head, the others short ; length of base of anal fin If-li times in
that of the base of the soft dorsal fin ; pelvis between vential
spines rather narrow, tapeiing posteriorly to a slender point.
Alembrane between first and second rays of spinous doi'sal fin
blackish.

Total length 180 mm.

Hah. East Indian Archipelago ; Australia ; China.

Triacanthus oxycephalus.

Triacanthus oxycephalus Bleeker, Yerh. Bat. Gen. xxiv. 1852,
p. 27, pi. v. fig. 10 ; Atlas Ichth. v. p. 90, pi. ccxx. fig. 3 (1865).

Triacanthus hiaculeatus {'^•Avt.)(3i\x\\t\\eY, Cat. viii. p. 210 (1870).

Depth of body 2y— 2| times in total length, length of head
about 3 times, length of caudal peduncle 6^—7 times. Snout
slightly concave, its length about 1-i times in the length of head,
eye-diameter 3-4 times, interorbital width 3-3j times. Inter-
orbital space flat. Upper edge of occipital crest convex ; distance
from postei'ior margin of orbit to base of first dorsal spine about
1| times as long as the eye-diameter. D. V, 24-25 ; A. 17-19 ;
first dorsal spine longer than the head, the others short ; length of
base of anal fin 14-1 1 times in that of the base of the soft dorsal
fin ; pelvis between the ventral spines broad anteriorly, tapeiing to
a point posteriorly. Membrane of spinous dorsal fin immaculate.

Total length 140 mm.

Hah. East Indian Archipelago.

Triacanthus brevirostris.

Triacanthus hrevirostris Schlegel, Faun. Japon., Poiss. p. 294,
pi. cxxix. fig. 2 (1846); Bleeker, Atlas Ichth. p. 94, pi. ccxxxi.
fig. 3 (1865); Gunther, Cat. viii. p. 209, part. (1870).

Triacanthus hiaculeatus Bleekei', Yerh. Bat. Gen. xxii. 1849, p. 6.

Triacanthus rhodopterus Bleeker, Yerh. Bat. Gen. xxiv. 1852,
p. 25, pi. iv. fig. 8.

Triacanthus russellii Bleeker, t. c. p. 25.

Depth of body 21 -2| times in the total length, length of head
3^-31 times, length of caudal peduncle 41-54 times. Snout
straight, its length about twice in that of the head, eye-diameter
3|-5 tijEces, intei-orbital width about '6^ times. Length of post-

184 ON THE FISHES OF THE GENUS TRIACANTHUS. [Feb. 17,

orbital part of head equal to or greater than eye-diameter. Inter-
orbital space with a more or less distinct median ridge, with a
groove on each side of it. Occipital crest convex, moderately
elevated, becoming nearly horizontal in front of the base of the
first dorsal spine. D. V, 24-25; A. 18-20; first dorsal spine
shorter than the head, the others short ; length of base of anal fin
1|— 1|- times in that of soft dorsal fin ; pelvis between ventral
spines broad, scarcely narrowed posteriorly. Membrane of spinous
dorsal fin blackish.

Total length 250 mm.

Hob. East Indian Archipelago, Seas of China and Japan.

Triacanthus indicus, n. sp.

Russell, Indian Fishes, p. 14, pi. xxi. (1803).

Balistes hiacideatus Bennett, Fishes of Ceylon, pi. xv. (1830).

Triaccmihus hiaculeatus Cantor, Mai. Fish. p. 360 (1847); Day,
Fishes of Malabar, p. 260 (1865).

Tricicanthus hrevirostris HoUard, Ann. Sci. ISTat. (4) i. 1854, p. 45,
pi. ii. fiff. 1 ; Giinther, Cat. viii. p. 209, part. (1870) ; Day, Fishes
of India, p. 685, pi. clxxv. fig. 1 (1878).

Depth of body 2^2 1 times in the total length, length of head
3|-3| times, length of caudal peduncle 4|-5 times. Snout
straight, its length about 1| times in that of the head, eye-diameter
3|-4| times, interorbital width 3-3| times. Length of post-
orbital part of head distinctly less than eye-diameter. Interorbital
space with a more or less distinct median I'idge with a gi'oove on
each side of it. Occipital crest elevated, its upper edge nearly in
the same straight line with that of the snout. D. Y, 24-25 ;
A. 18-20 ; first dorsal spine nearly as long as the head, the others
short ; length of base of anal fin 1^-1 1 times in that of soft dorsal
fin ; pelvis between ventral spines bi'oad, scarcely nai-rowed
posterioi'ly. Membrane of spinous dorsal fin blackish.

Total length 220 mm.

Hah. Coasts of India from Kurrachee to Penang ; Ceylon ;
Andaman Is.

This species difiers from T. hrevirostris chiefly in the longer
snout, the shortei' postorbital pait of head, and the shoi-ter and
more elevated occipital crest. The figures given by Russell and
Bennett are excellent.

Triacanthus nieuhofi.

■^ Nieuhof, Gedenkw. Zee en lantr. p. 272, fig.

Willoughby, Ichthyology, Appendix, p. 5, pi. x. fig. 2 (1686).

? Gronow, Mus. i. p. 52 (1754-6) & ZoophyL p. 53 (1763-81).

Triaca7ithics 7iietchofi Uleeker, Verb. Bat. Gen. xxiv. 1852, p. 26,
pi. iv. fig. 9, & Atlas Ichth. v. p. 92, pi. ccxvii. fig. 3 (1865).

Triacanthus hrachysoma Bleeker, Nat. Tijds. Ned. Ind. iv.
1853, p. 128.

1 I have not hcew nble to verily tliis reference.

1903.] ON VAUIATIOXS OF THE SAXD-VIPER. 185

"i Balistes hipes Gronow, Cat. Fish. p. 37 (1854).

Triacanthushrevirostris (part.) Giinther, Cat. viii. p. 209 (1870),

Depth of body about 2^ times in the total length, length of
head 3g-3|- times, length of caudal peduncle ^l^— 4^ times. Snout
straight, about 24- times in the length of head ; eye-diameter about
3j times and equal to interorbital Avidth. liOngth of postorbital
part of head less than eye-diameter. Inteioi'bital space with
median ridge with a groove on each side of it ; occipital crest
strongly elevated, the distance fi'om the base of fii'st dorsal spine
to the upper angle of gill-opening rather more than the distance
fi-om the base of the fii'st dorsal spine to the anteiior margin of
eye. D. V, 24-25 ; A. 18-20; first dorsal spine longer than the
head, the others shoit ; length of base of anal fin about 1 h times
in that of the base of soft doi-sal fin ; pelvis between the venti'al
spines broad, not much narrowed posteiiorly ; membi*ane between
first two rays of spinous dorsal fin blackish.

Total length 126 mm.

Hah. East Indian Archipelago.

This species has a shorter and more declivous snout than
Tr. hrevirostris, and also occipital crest more elevated, postoi'bital
part of head shorter and body deeper.

5. On the Geographical Variations o£ the Sand-Yiper, Vipera
amuwdytes. By G. A. BouLENGER, F.ll.S., V.F.Z.S.

[Received January 30, 1903.]
(Text-figures 27 & 28.)

The variations of Vijyei-a ammodi/ies in connection with the
distribution of the sjiecies have not I'eceived sxifiicient attention.
Having succeeded in bringing together and carefully compaiing a
series of 55 specimens from various localities, I have convinced
myself that the South-eastern specimens (Gi'eece, Archipelago,
Syria) can be distinguished from the typical form from Austi-ia-
Hungaiy, Dalmatia., Bosnia, and Montenegi'o, not by means of
any single absolute character, but by a combination of characters,
as shown by the following definitions : —

Forma tijpica (text-fig. 27 a). — Naso-rosti'al shield usually ^
reaching the canthus rostralis, and extending considerably higher
up than the upper border of the I'ostral, which is usually broader
than deep (text-fig. 28 a) ; rostral " horn " with 3 (rai'ely 2 or 4)
transverse seiies of scales between the rostral shield and the
apex. Ventral shields 145 to 163. The dark shade on the lower
lip, if present, broken up by liglit bars sepai'ated by 2 to 4 labial
shields. Lower surface of end of tail usually red ^. Grows to
80 centimetres.

Var. meridlonaUs (text-fig. 27 b). — ISTaso-rostral shield never

1 5 exceptions out of 30 specimens examined.

" Yellow in one specimen from the Dinaiic Alps', Bosnia.

186

ON VARIATIONS OF THE SAND-VIPER.

[Feb. l\

reaching the canthns rostralis, and hut rarely extending higher
tip than the upper border of the rostral, which is often as deep as
broad or a Httle deeper than broad (text-fig. 28 b) ; rostral " horn"
with 3 to 5 transverse series of scales between the rostral shield
and the apex. Supraciliary edge usually more prominent than in
the typical form, sometimes slightly angular, foi'eshadowing the
condition in V. raddii. Ventral shields 133 to 147. A more or
less distinct dark blotch on the lower lip, involving 5 or 6 labial
shields without interruption. Lower surface of end of tail yellow.
Grows to 60 centimeti-es.

Text-fig. 27.

Side views of heads of a, Vipera ammoch/tcs, f. ti/pica (Feldkirchen,
Garinthia), and b, var. meridionalis (Athens).

Text-fig. 28.

D

Front view of end of snout, showing the lepidosis.
a. Fehikirchcn, Garinthia ; h. Athens ; c. Cocosu, Roumania.

I submit these definitions to the consideration of herpetologists
who may have the privilege of examining large series of speci-
mens from any single district. The material at my command is
still deficient in examples from Bulgaria, Turkey, and the Caucasus,
and I may mention that the only two specimens from Roumania
which I have been able to examine differ from both forms as here
chai'acterized, and are on the whole intermediate between them.
The rostral shield is deeper than broad, and the naso-rostral does
not reach the canthus rostralis (text-fig. 28 c) ; one has two series
of scales on the rostral " horn," the other has three ; supraciliary
edge not very prominent ; ventral shields 155 and 153 ; no light
bars on the lower lip ; lower surface of tail yellow.

1903.] ON THE HABITS OF THE HOOLOCK. 187

G. Notes on the Habits of the Hoolock.
By Geo. Candler, M.B.Cantab. ^

[Received January 20, 1903.]

The Hoolock {Ilijlolates hoolock) is one of the most interesting
of the family Simiidte, and is perhaps not so familiar to natni'alists
at home as are some other members of the family, as, owing to its
extreme delicacy and the great ditiiculty experienced in keeping it
alive in confinement, it does not often find its way into European
collections. Even in the Calcutta Zoological Gardens it is difficult
to keep Iloolocks alive for any length of time. They often
succumb to pneumonia, or if they escape actual disease they mope
and die from the eftects of confinement, or possibly from depri-
vation of some article of diet Avhich in the wild state they have
been accustomed to. I have ventured therefore to submit to the
Society these short notes, made fi-om the point of view of a field-
naturalist luther than fi-om a scientific aspect.

The Hoolock is clothed all over with a fine soft hair, which,
perfectly black in the male, in the female shows a greyish tint,
especially over the back. This uniform dark colour is only
relieved by narrow horizontal streaks of white hair above the eyes.
The face, palms of the hands, and soles of the feet are devoi(l of
hair, and here the black skin is smooth and finely wrinkled and as
soft as the finest kid.

The hallux and pollex have a flattened nail, the remaining digits
have the nail laterally compressed and resembling a claw.

Thei'e is no tail. Ischial tuberosities and cheek-pouches are
absent.

Wlien the Ape is sitting, the vertebral column presents a single
marked curve with the convexity backwai-ds. On the ground the
Hoolock has a very characteristic gait. He goes along in a sort of
sliambling waddle, with legs bowed and knees bent, the soles of his
feet applied flat to the ground with the hallux widely abducted, both
arms being carried upwards and extremely alxlucted as if to balance
himself. He cannot get up any speed, and invarialfly swings up
into the first tree he comes to, where his movements are suddenly
changed from extreme awkwardness to extiuoi'dinary grace and
agility.

He swings along to the thinnest part of a liough, or to the
slender end of a bamboo, until it bends to his weight, then with a
swing and a soi't of a kick-off he flies through the air, seizing
another bough and swinging along it with the unerring accuracy of
a finished trapeze performer. I fancy he does very little walking
in the wild state, for I have never seen a wild Hoolock on the
ground. Moreover, they are only found in the dense jungle where
the ground is everywhere covered by tangled vegetation. It is
puzzling to me why these anthropoids, being so entirely arboreal

^ Coiiinmiiicatc'J liy F. (!. Paiisons, F.Z.S.

188 MR. G. CANDLER ON THE [Feb. 17,

in habit, should be lacking in such a useful appendage as a tail,
I think, at any rate, that it points to the fact that the apes have
been developed along a line distinct from the monkeys, the earlier
traces of which line are yet to be discovered.

The Hoolocks are extremely shy, and it is most difficult to
watch them, as they are concealed by leaves high up in the tops of
the bamboo- clumps or foi'est trees. You may hear their cries all
round you as you I'ide quickly along a jungle-tract, but the
moment you leave the path or look up at them there is a dead
silence and scarcely a leaf stirs, until, tii'ed of waiting, you move on
again.

The cry of the Hoolock is a characteristic sound in the Cachar
jungle. It is a very pleasing note, rising and falling in intensity,
and reminding one somewhat in its rhythm of a pack of beagles
giving tongue on a scent which is waxing and waning in strength,
as a larger or smaller number of the band join in the choiais.
It is heard chiefly in the early morning, then all through the heat
of the day there is silence, but towards evening, as the sun sinks,
you may hear it again. Hooloo ! Hooloo ! Hooloo ! with the accent
on the Hoo syllable, is supposed to describe the sound, but it is
really quite indescribable in writing.

As in other species of apes, there is a special modification of
the larynx, which acts as a sort of i-esonating-box, and helps (I
suppose) to make the sound carry, as it does, long distances. There
is also a peculiar arrangement of the upper aperture of the larynx,
with its small and inadequate looking epiglottis, which more
resembles the arrangement in birds than the leaf-like epiglottis in
man.

As, day after day, I have I'idden through the jungle, it has seemed
to me that the Hoolocks work their ground systematically in
their search for food, just as the planter plucks one section of
his tea to-day and another section on a distant part of the garden
to-morrow. Foi' I have found them filling the air with their cries
along a particular sti'etch of jungle-road one day, whilst the next
day not one was to be heard ; then, perhaps, a week later they ai'e
back again in the same place. Living as they do in communities,
they ai'e constantly on the move, and from what we know of their
great intelligence, it seems to me highly probable that their move-
ments are guided by very definite plans, and that very probably
they have some sort of government system.

There is a point about the Hoolock that strikes me as very
extraoi'dinary, and that is the fact that he cannot swim. I had been
told this by both natives and Europeans, but I confess I was
somewhat sceptical about it until I tried experiments myself. We
put a full-grown Hoolock into a big tank in 10 feet of water. He
struggled helplessly, as a boy would before he learns to swim.
He sank twice, with head thrown back and arms waving franti-
cally, and we were obliged to rescue him almost asphyxiated and
choking in the most human way.

1903.] HABITS OF THE HOOLOCK. 189

This weakness he shaves witli man, but I do not know whether
(or not) it has been noted in the otlier antliropoids.

It is a significant fact that the range of the Hoolock is l)ounded
by two vast rivers, tlie Bralmiaputiu on tlie north and the Irawaddi
on the soutli. It may well he that, with his natural aversion to
water, these rivers have confined hiin to the comparatively limited
stretch of countiy lie occupies. Tiuvelling high up in the jungle,
he could swing easily across the ordinaiy sti'eanis which would
come in his path ^\•ithout having to take to tlie water. The
monkeys of India take readily to water, and it is a pretty sight to
see them spring out from a lofty overhanging bough and drop, one
after another, with a splash into the stream, and stiike out Ijoldly
for the further bank.

In Cachar, where these notes were written, the tea-plantei'S
often keep Hoolocks for years, allowing them to run loose about
the compound, and they are certainly the cleanest and most
interesting pets imaginable, ofiering a \evy marked conti'ast in
this respect to the red monkeys, which, chained to a pole, are so
common a feature in Indian compounds.

A Hoolock, to be tamed in this way, must be caught quite young,
and not tied or shut up in any way. A native l)oy is generally told
ofl:' to watch him for a few days, and to prevent him from bolting,
but he soon learns to come down from the ti'ees for a plantain,
and he will in most cases settle down to a solitary life, I'emaining
about the .same compound for years. But chain him or restrain
his libei'ty in any way, and he inevitably begins to mope and pine,
and invai'iably dies in a, few weeks. It is strange that the calls of
the wild Hoolocks, which he must hear almost daily all round him,
do not tempt him to revert to his natui-al life as a member of a
wandering community. I imagine a Hoolock, who attempted to join
a strange band, would meet with a rough reception, anyway they
never try to return to the jungle after they become tame.

Several such tame Hoolocks I have had the opjioitunity of
observing for some months past. Often they will be away up in
the tree-to}is for days together, when nothing will tempt tliem
down, ]:)ut when one chooses to be sociable he will come and sit
on the arm of your chair at breakfast, and never reach or snatch
things oft" the table: in fact his manners are unexceptionable, and
he keeps his skin beautifully clean without that exaggerated
parade of flea-hunting which makes the monkey tribe so oi>jection-
able as pets. At sunset a-ou may see him settle down to sleep,
jammed tight in a fork in a squatting position. In this semi-
domesticated state I notice that the Hoolock seldom uses his
voice. I suppose, leading a solitary bachelor life, he finds no
necessity for chattering or calling. "With regard to the diet of
the Hoolock, Dr. Blanford, the Indian natui'alist, gives a long
list, including fruit, leaves, young shoots, spiders, insects, birds'
eggs and young birds. But, it seems to me, the diet of such shy
creatures must be largely a matter of conjecture, for no certain

190 ON THE HABITS OF THE HOOLOCK. [Feb. 17,

conclusions can be drawn from the habits of captured specimens,
nor can we recognize as a rule substances in the stomach of shot
specimens, as we can in the crop in the case of birds. My own
observations lead me to believe that fruits and the succulent shoots
of young bamboos and other trees form the bulk of their diet.
They will certainly catch and eat certain spiders ; but I have
invariably found them to refuse such insects as moths or butter-
flies, perhaps because many such insects have a bitter taste. Eggs,
too, I found they would not eat. If you give an insect or a small
bird to a Hoolock he will certainly pull it to pieces, and possibly
taste or bite it, but it by no means follows that it is one of the
regular dishes he enjoys in his wild life.

The following list of leaves and shoots which are eaten by the
Hoolock is given by Anderson: — Morugo pterygosperitna, Sj^ondias
manifera, jFicus religiosa, Beta vulgaris, Ipomcea reptans, Canna
indica.

I hope later on to supplement these short notes with some
anatomical observations on weight of brain relative to body, and on
the number and depth of the convolutions. But this is a matter
of time, for specimens are not very readily obtained. The Hindoo
coolies, who form the bulk of the population in the tea-districts
of Cachar, will never kill a Hoolock. The Kuki tribes in the
Oachar Hills, on the other hand, kill and eat them, and regard
them as somewhat of a delicacy, I believe. But even a Kuki finds
it diificult to get a shot at these creatures, so shy are they and so
active in their movements.

1903.] ox CERCOCEBUS ATERKIMUS AXD LUTRA CAPEXSIS. 191

March .'{, 1903.

G. A. BouLENGER, Esq., F.R.S., Vice-President,
in the (Jhair.

The Secretary read the following report on the additions made
to the Society's Menagerie in Fel)ruary 1903 : —

The registered additions to the Society's Menagerie during the
month of February were 57 in ninnber. Of these 8 were acquired
by piesentation, 48 were received on deposit and 1 on approval.
Tlie total numbei' of departures dui'ing the same period, by death
and removals, was 105.

Amongst the additions special attention may be directed to : —

1. An example of Ouvier's Clazelle {Gazella citvieri), deposited
by the Hon. Walter Rothschild, M.P., F.Z.S., on Feb. 9th. No
example of this i-are North- Afiican Gazelle has been exhibited in
the Gaidens since 18G7.

2. A Tamandua Anteater [Taviandua tetradactyla) from South
Ameiioa, received on approval on Feb. 12th. This is a very
health}' and lively specimen of an animal which we have not had
living in the Gardens for some yeai'S.

3. A young nrde Chimpanzee {Anthropopithecus troylodi/tea)
about a year old, deposited by Mr. J. C Lamprey, of the West
African Regiment, on Feb. 18th. This animal is said to have
been obtained at Kionko in the French Soudan, and was brought
home from Sieriu Leone by the depositor. It is in very good
health.

4. A Fiilletl Lizai'd (C/damydosaui'us kingi) from Australia is
the second example of this species received at the Gardens, the
first having been pi-esented by Mr. Saville Kent some seven years
ago. The piesent specimen was presented by IMi-. H. W. Fawdon
on February 18th.

The Secretary read the following extracts from a letter addressed
to Mr. P. L. Sclater by Major C. Delme Radclifi'e from Uganda,
and exhibited the skins of a Monkey {Cercocelms aterrhims) and
an Otter {Lutra capensls) obtained by his collector J\Ir. Doggett,
and sent home by parcel post, at the same time : —

" I am sending you the skins of two monkeys — one completely
prepared with bones, and the skin of the other, also skulls of both.
The animals were both female. Doggett got them for me a few
days ago, as I had sent him collecting by road here from Entebbe.
This monkey is quite new to me. I shall be glad to hear from you
if it is a new species. It is very intei-esting, and the skull has .some
interesting features, for instance the small canines. The hair is
curiously like a Chimpanzee's. Doggett tells me he thought from
the noise they made that they were Chimpanzees at first. Is it a
true Cercopithecus '? It interests me very much, and I shall be

Proo. Zool. Soc— 1903, Vol. I. No. XIII. 13

192 ME. w. E. DE wiNTON OX [Mar, 3,

glad of your opinion. In the same parcel I enclose a Lake Otter
skin."

Mr. F. E. Beddard, F.K.S., exhibited the mounted skin of the
Greater Bird of Paradise [Paradisea apoda) that had lately died
in the Society' Gardens.

Mr, J. L. Bonhote, F.Z.S., exhibited a photograph of two tame
Elephants in Ceylon accompanied by a baby one, which latter
showed a considerable amount of hair, especially on the forehead.
One of the old Elephants in the photograph, presumably the
mother, also showed a ceitain amount of hair.

With reference to a recent paper by Col. C. E. Stewart, dealing
with the real home of the Tiger, Mr. J. L. Bonhote said that he
had asked a friend of his, a Sanskrit scholar at Cambridge, as to
whether there was a word foi' the tiger in Sanskrit ; the reply was
in the aliirmative, the word being vydghra (or in the Pali form,
vyaggho). This note was not brought forward against the fact
of the original home of the tiger being in the north, but to support
Mr. Thomas's contention that it had not spread south so recently
as Col. Stewart was inclined to believe.

Professor F. Jeffrey Bell, F.Z.S., exhibited a specimen of a
Holothurian of the genus Actinopyga from shallow water off
Zanzibar, in which there was not only an oral extremity with
tentacles, as in the Ctocumaria planci described some ten years
since by Prof. Ludwig, but also an anal extremity ; these additions
(whether the result of gemmation or of fission cannot at present
be said) do not occur in the same radius of the Holothurian.

The following papers were read : —

1. On a new Species of Pigmj^ Antelope of the Genus
Neotragus from the Cameroons District^ W. Africa.
By W. E. DE WiNTON, F.Z.S.

[Received March 3, 1903.]

(Plate XIX. & Text-fig. 29.)

In a collection lately received from Mr. G. L. Bates from the
Cameroons are two specimens, male and female, of a new form of
Pigmy Antelope, adding a second species to the genus Neotragus.
I propose to name this very interesting new form in honour of the
collector, who has added so much to our knowledge of the fauna
of the Camei'oons country.

X

en
o

OQ

H
O

1903.]

A NEW PIGMY AXTELOPE.

193

Neotragus batesi, .sp. n. (Plate XIX.)

Size nearly half as large again as the Royal Antelope, jV. pyy-
mceus. The colour ilarker and slightly more smoky, the feet not
.so rufou.s, and the wliitish markings of the uudei-pai-ts not so
puie.

The front of the face, crown of the head, ami neck are dark
smoky- hi-own, the bnck and upperside of tail more rufous or bay-
coloured. All the fur annulated with dark brown and bright bay.

The t^vpe of this new species, a male, bears the following notes
on the label attached by the collector: — " Efulen, Bulu Country,
Kamai'un, 1500 ft. above sea. Caught in trap on edge of garden,
15 Sept. 1902." Measureuieiits : head and Ijody 540 mm., tail 65,
hind foot (without hoof) IGO, ear (inside from notch) 50.

Dimensions of skull of male (text-fig. 29): — Greatest length
107 mm.; basal length 95; greatest breadth 51 (below orbits) ;
bi'eadth of brain-case 37'5 ; orbit to points of premaxilla? 48 ;
length of nasals (middle line) 33 ; length of frontals 33 ; length
of parietals 30; length of interparietals 13; greatest breadth of

Text-fio-. 29.

Skull of Neotrac/ us hate

nasals 20 ; breadth in narrowest pai't 9 ; lengtli of horn (biolien)
27, the peifect horn could not be less than 35 mm. in length ;
distance between bases of horns 25 ; distance between tips
{circa) 35 ; length of orbit 23 ; height of orbit 22 ; length of
upper tooth-row 32; breadth of palate between pms.' 17, between
ms.' 20, between ms.^ 21 '5.

The dimensions of the skull of the female are practically the
same as those of the male.

The discovery of this new species extends the I'ange of the genus
Xeotragus very considerably, for the Royal Antelope (i\\ pyymceus)
is found onlv on the Gold Coast from Liberia to Lagos.

^13*

194 MR. E. R. SYKES ON [Mar. 3,

An examination of the skull somewhat modifies the definition
of the genus as laid down in ' The Book of Antelopes ' by Sclater
and Thomas ; for in this new species are found extra maxillo-
premaxillary vacuities 9 mm. long and 3 mm. broad, very similar
to those found in the genus Nesotragus of East Africa. The
horns again of the members of this genus cannot be said to be
" less than the diameter of the eye " in leng-th ; for the horns
of the type of this new species, in their broken state, are longer
than the diameter of the orbital cavity, and would undoubtedly
measure half as much again in their perfect state. These horns
are not "perfectly smooth," but show slight ridges or rings in
their basal poi-tion. The skull is otherwise fairly similar to that
of the Royal Antelope.

There can be no doubt that Bates' Pigmy Antelope somewhat
bridges over the difierences between this genus and the East-
African Nesotragus, and practically reduces the distinguishing-
characters to those of the horns alone.

The horns of Neotragus are very small, practically smooth, and
laid back on the head in a plane with the forehead ; while those
of Nesotragus may be half as long as the head or more, strongly
and closely ridged and directed upwards.

2. On the Land Operculate Mollusca collected during the
" Skeat Expedition " to the Malay Peninsula in 1899-
1900. By E. R. Sykes, F.Z.S.

[Received February 2, 1903.]

(Plate XX/)

The species of Land-Mollusks collected by the "Skeat Expedi-
tion," though not very numerous, are of considerable interest on
account of the welcome addition made to a fauna which is as yet
but little known. They include : —

Leptopoma aspirans Benson.

Leptopoina aspirans Benson, Ann. Nat. Hist. ser. 2, vol. xvii.
p. 229.

Hah. Biserat, Jaloi-. It has been recoided from Bukit
Pondong and the Kinta Valley, having been originally described
from Tenasserim.

Lagochilus kobelti, sp. nov. (Plate XX. figs. 13-15.)
Testa modice tiinbilicata, turhinata, solidula, lirulis numerosis
cincta, periostraco brunneo leviter induta ; spnra conica ;
anfr. 5j, convexi; aper tur a modice ohliqua, suhcircularis, peri-
stomate dtiplici, incrassatulo et suhreflexo, juxta insertionem
breviter sed distincte inciso. Alt. 6"8 ; diam. max. 7'0 millim.
Hab. Biserat, Jalor.

1 For explanation of the Plate, see p. 199.

P.Z.S.1903,vol.I.Pl.XX.

J.GT-eeTxael.etifK 18. 19. Mi:r.terix Bros imp.

MOLLUSCA FROM THE MALAY PENINSULA.

1903.] MOLLUSCA OF THE "SK EAT EXPEDITION." 195

The species of Lagochilus recorded from the Malay region are
very puzzling and I think Dr. Moellendorft" Wiis qixite right in
describing as L. rollei the form that I noted (under the name of
L. toivnsendi) from Kelantan. The nearest ally of L. kohelti
appears to be L. toivnsendi Crosse ; I have not seen an authentic
specimen of that species, but have compared the form now described
with the description and figures given by Crosse and with a
specimen collected by Herr Grubauer, from whose collections
Dr. Moellendorft" has recorded L. tovmsendi as the only species
foinid. The shell I now name is a trifle smaller and more
elevated in proportion to the breadth, and the umbilicus is
narrower. It may be noted that the reference to Crosse's original
paper in the ' Journal de Conchyliologie ' should be to p. 200 and
not p. 208 as given by De Morgan and Dr. Moellendorff in their
papers on the Pei'ak fauna.

I have named the form after Dr. Kobelt as a trifling recognition
of his recent study of the Cyclophoiida?.

DiTROPis CAVERN^E, sp. nov. (Plate XX. figs. 17-19.)

Testa depresso-conoidea, late tombilicata, olivacea, tenuis, glabra ;
s]nra medioa'iter elevata, apice eroso, sutiiji'a impressa ;
anfr. 4 (?), convexi, ulthmis antice vix descendens, carinatus,
carinis chmhus su])ra pei'iphe7-iavi, unica ad, jjeriphet-iain, et
plurimis in regione umbilicali ; apertura subovalis, peristotnate
incrassatido. Alt. 2*2 ; diam. max. 1*7 milliiti.

Hab. In a cave, Biserat, Jalor.

A single specimen only.

Cyclophorus malayanus (Benson).

Cyclostoma malai/anitm Benson, Ann. Nat. Hist. ser. 2, vol. x,
p. 269.

JIab. Gunong Inas, Perak.

Recently when cataloguing (J. Malac. ix. p. 61) a collection of
shells from Kelantan, I gave the names of Cyclojyhorus saturnus Pfr.
and borneensis Mete. : both the forms then recorded have occurred
in the present collection, and I have therefore again considered
the identifications. Both, according to my present view, are
erroneous, and the group is a very difficult one. The present
species, which I regard as a form of C. malayanits, was then named
G. saturnus, and the next species was called C. borneensis.

Cyclophorus tuba (Sby.).

Cyclostoma f^66rt Sowerby, Pi'oc. Zool. 8oc. 1842, p. 83.
Hab. Gunong Inas, Perak.

See remarks under the last species ; probably the C borneensis,
recorded by De Morgan from Perak, also belongs to this species.

Pterocyclos subalatus, sp. nov. (Plate XX. figs. 1, 2.)
Testa late umbilicata, orbiculato-dejiressa, lineis incrementi notata,

196 ME. E. R. SYKES OK [Mar. 3,

hrunnea, strigis cornels picta, fascia unica nigro-hrunnea ad
peripherimn ornata ; anfr. 4|, mediocriter crescentes, coiivexi,
sut'iora bene notata separati ; apertura siobcircularis, peri-
stomate indistincie duplicato, ala parva antice aiigidato.
Alt. 8 ; diarn. max. 1 6 millim.
Hah. Gunong Inas, at 5000 feet,

I thought at fii\st that this might be a form of P. hlandA Bens.,
but it appeal's to be smallei', more elevate and not so widely
umbilicated, and to difler in colour. In the two specimens that I
have seen the lip is diiplicated only on its outer mai-gin, and the
wing is small and thin.

Rhiostoma jalorensis, sp. nov. (Plate XX. iigs. 6-8.)

Nearly related to R. housei but larger, slightly more depressed,
and with the whorls not so tightly coiled. The tube is long and
large, reaching the body of the shell, and is bent slightly back-
wards at the junction. The disjoined poition of the last whorl is
much longei' than is the case in E. housei, as will be seen from
the figure, and is more descending. The colour resembles that
of B. housei, and a dark band is usually present at the periphery.
Operculum as usual. Diam. max. 29 millim.

JIab. Limestone Hills and Oaves, Biseiat, Jalor.

I have been in some doubt as to whether this is not a local race
of 7?. housei, but the difierences are constant in the specimens
examined and, I think, are of specific value. The porcellaneous
appearance of the operculum, so often seen in this group, appears
to be due to the wearing down of the outei- layers.

Rhiostoma, sp.

Hah. Kwala Aring, Kelantan.

A single specimen which agrees well with R. jousseaumei
De Morgan, save that the tube is placed a little further back from
the mouth, and is thinner and moi'e cylindrical. It appears safer,
however, to await fui'ther material ere describing this form.

Opisthoporus penangensis Stol.

Opisthoporus penangensis Stoliczka, J. As. Soc. Bengal, vol. xlii.
pt. 2, p. 265, pi. X. fig. 7.

Hah. Kwala Aring, Kelantan.

A single specimen, which I refer to this species with some doubt.

Rhaphaulus ascenders, sp. nov. (Plate XX. figs. 11, 12.)

Testa pupoidea, angxiste ^mnhilicata, fusco-purp)urea, dense costu-
lato-striata, sp>ira bene elevata, apice obtuszdo ; anfr. 6, plano-
convexi, p)emdti')nus -gibbosus ; apertura suhcircidaris, intus
pallidefusca; peristoma jxdlide corneum, exjjansitm et re/lexum,
marginihus callo junctis ; tuhulus brevis, incrassatus, a sutidra
oblique ascendens. Alt. 18'5; diam. max, 9*5 millim.

Hah. Patalung. ,

1903.] MOLLUSCA OF THE " SKEAT EXPEDITION." 197

A single specimen "fi-om rotten wood." Recalling in shape
R. jjerakeiisis )Smith, the pi'esent species is larger and stouter, and
the tube slants obliquely upwards instead of descending ; as com-
pared with R. lorraini Pfr., which also has an ascending tube, the
form is not so cylindrical, the whorls are flatter, and the size is
much greatei'.

Rhaphaulus perakensis Smith, var. jalorensis, n. var.
(Plate XX. figs. 9, 10.)

Shell very similai' to R. perakensis Smith, but a little thinner
and palei- in colour, and the tube is bent along the outer lip, being
attached to the lip, a.nd not diverted behind it as in typical
R. perakensis. The tube is broad and short. This form may pi'ove
to be a distinct species, but bearing in mind the great vai'iation
shown by Col. Godwin-Austen (Moll. India, vol. i. p. 202, pi. xlvii.)
to exist in R. blavfordi Benson, and also considering how little we
know of the group, most forms having been described from very few
specimens, I have thought it wiser to give only a, vai'ietal name.

Hab. Bukit Bisar, on the borders of Jalor, altitude 2000 feet.

A single specimen.

Schistosoma anostoma (Benson).

Cyclostoma anostoma Benson, Ann. Nat. Hist. ser. 6, vol. x.
p. 269.

Uab. Belimbing, Ligeh.

SCHISTOLOMA SECTILABEUM (Gould).

Cyclostoma sectilabriim Gould, Boston Journ. Nat. Hist, vol, iv.
p. 459, pi. xxiv. fig. 10.
Hab. Ulu Selama, Perak.

PuPiNA Lowi De Morgan.

Pupina loivi De Morgan, BuU. Soc. Zool. France, vol. x. 1885,
p. 414, pi. vii. fig. 3 {loitri on plate).
Hab. Gunong Inas, Perak.

PuPiNA AUREOLA StoHczka.

Pupina aureola StoHczka, J. As. Soc. Bengal, vol. xli. pt. 2,
p. 267, pi. X. figs. 11, 12.

Hab. Jalor, a single specimen.

Agrees well with Stoliczka's figure 1 2, but his figure 1 1 looks as
if it might belong to a different species.

Alyc^us thieroti De Morgan.

Alycceus thieroti De Morgan, Bull. Soc. Zool. France, vol. x.
1885, p. 403, pi. viii. fig. 6; Moellendorfl; Proc. Zool. Soc. 1891,
p. 342.

Hab. Belimbing, Ligeh, a single specimen.

198 ON MOLLUSCA OF THE " SKEAT EXPEDITION." [Mar. 3,

Alyc^us diplochilus Moellendorff.

AlyccBus diplochilus Moellendorff, J. As. Soc. Bengal, vol. Iv.
pt. 2, 1886, p. 313 ; Proc. Zool. Soc. 1891, p. 342, pi. xxx. fig. 8.
Hah. Cave near Biserat, Jalor.

AlyCjEus conformis Fulton.

Alycceus covformis Fulton, Ann. Nat. Hist. ser. 7, vol. ix. 1902,
p. 68.

Hab. Kwala Aring, Kelantan.

Alyc^us perakensis Crosse.

Alycceus perakeoisis Crosse, J. Conchyl. vol. xxvii. 1879, p. 206,
pi. xii. fig. 7.

Hab. Biserat, Jalor, amongst rocks.

Opisthostoma annandalei, sp. nov. (Plate XX. figs. 4, 5.)

Testa conico-pyramidalis, perforata, ])allide riifescens, spira bene
elevata, apice obtuso ; anfr. 7, convexi, primi Iceves, reliqui
distante lamellati, ultimios constrictus, retroversus, ascendens,
conspicue solutus ; apertura rotundata, peristomate leviter
incrassato. Alt. 2"2 ; diam. max. 2"8; diam. niin. 1*6 millim.
Hab. Jalor.

A single specimen, found by Mr. Annandale in debris on the
floor of a cave. Related to 0. laidlawi Sykes, from Kelantan, but
the pi'esent species is larger, and the spire is much more produced,
thus becoming more cylindrical in appearance.

DiPLOMMATINA SKEATI, sp. nov. (Plate XX. fig. 3.)

Testa sinistrorsa, vix rimata, ovato-fusiformis, aurantio-rufa,
solidiuscida ; anfr. 6^, convexi, p7-imi loives, reliqui costidis
p)arvis subremotis regidariter scidpti ; sutura bene impressa ;
apertura fere circtdaris, peristomate incrassatido, bene reflexo,
subalato ; lamella columellaris paj-va. Alt. 3*1; diam. max.
r9 millim.
Hab. Gunong Inas, Perak, at about 5000 feet.
Only a single specimen of the shell, the most salient features of
which are the inflated whorls, deeply cut suture, and fine, regular
costulations.

DiPLOMMATINA LAIDLAWI, sp. nov. (Plate XX. fig. 16.)

Testa sinistrorsa, rimata, ovato-fusiformis, cornea- albida, apice
ritbello ; anfr. 5|, convexi, primi Iceves, reliqid costulis parvis
remotis regidariter scidpti, interstitiis sub lente dense spiraliter
striatis ; apertura quadrato-ch-cidaris, j^^^i^iomate duplici,
expanso, margine coluinellari sinuato, subalato; lamella
columellaris minima, indistincta. Alt. 2; diam. max.
1 millim.

Hab. Gunong Inas, 3000-4500 feet, Perak.

Found " on the under surface of dead leaves, young with adults."

1903.] ox THE CALLOSITIES OX TME LIMBS OF THE EQUID.E. 109

Much smaller than the last species, the costula^ more distant, and
densely marked with minute spirals. I can find no trace of these
latter in T). skeati, hut the single specimen was not taken alive.

Georissa moxtekosatiaxa Godwin-Austen k. Nevill.

Georissa monterosaliaua Godwin- Austen & Nevill, P. Zool. Soc.
1879, p. 729, pi. lix. %. 6.
Hab. Pcrak.

EXPLANATION OF PLATE XX.

Figs. 1, 2. Fteroci/cJos suhalatus, p. 105.
3. Diplommatina skcati, p. 198.
4, 5. OpiaHioslnina annandalei, p. 198.
6,7,8. R/iiosfimiti Jalorensis, p. 196.

9, 10. Rhaphaidus perakfusis, var. Jalorensis, p. 197
11, 12. Rhaphauliis asccndons, p. 196.
13, 14, 15. Lngorhilm kohciti, p. 194.

16. Diplommatina laidlawi, p. 198.
17, 18, 19. Ditropis caver nee, p. 195.

3. The Significance of the Callosities on the Limbs of the
E<pdd(v. By R. Lydekker, F.Z.S.

[Received Februaiy 5, 1903.]

The question as to what structures in other mammals are
represented by the callosities on the inner sides of all the limbs of
the Horse, and those of the hind-lindis of the Kiang, Ass, and
Zebras, is one which has attracted the attention of many naturalists,
nearly all of whom appear to l)e in accord in i-egarding them as
vestigial structures. The late Sir W. H. Flower, for instance, in
his work ' The Horse ' expressed the opinion that these structures
ai'e degenerate glands ; pointing out at the same time that the
so-called ei'got on the hinder aspect of the horse's pastern appears
to represent one of the pads, or cushions, which aie still functional
in the foot of the Tapir.

In his volume on " Mammalia " in the ' Cambiidge Natural
History'', of which the preface is dated Feliruary 1902, Mr.
Beddard gives a qualilied support to this gland-theory ; stating in
one passage that the equine front callosities probably coirespond
to the carpal glands of several other mammals, although on
another page their glandular nature is questioned. In an
apparently later communication " the same gentleman suggests
that the callosities on the fore-lind)s of the EquAdjce may represent
a carjial sense-organ, vestiges of which he believes to survive in
the caipal bristles of the Dassies. The degeneration of such an
organ would, it is urged, very likely result in the formation of
sti'uctures resembling those under consideration.

1 Pp. 12, 13, and 210.

2 Proc. Zool. .Soc. 1902, i. p. 135. — I am indebted to Mi-. Heddard liimself for
directing my attention to thi.s passage.

200 MR. R. LYDEKKER OX THE CALLOSITIES [Mar. 3,

There has, however, long existed an idea that the equine callo-
sities are i-emnants of a vanished toe. Somewhat analogous to
this idea is a theoiy, which has been advanced in a papei* lately
read befora the Royal Society of Edinburgh by Prof. Ewart \
That genl:leman, it appears, has based his investigations to a
very great extent on the condition obtaining in the foetus ; and
has arrived at the conclusion that the callosities in question have
nothing to do with glands, but that they i-epresent certain foot-
pads of polydactyle mammals. According to Prof. Ewai't's view,
the foi'e- callosity of the Hoi'se is homologous with the supplemental
pad on the fore-foot of the Dog, while the hind callosity (which is
wanting in all existing Equidce save the true Horse and the so-
called Eqitus jjrsewalskii) corresponds to the hinder plantar pad
of the Banded Anteater (^Myrmecohiusfascicdus). Such correlations
will, I venture to think, scarcely be taken seriously by the great
majoiity of zoologists ; and I shall thei'efore confine my attention'
to the question whether the identification of these callosities with
foot-pads geneiully is well founded. In all cases my observations
are confined to the condition obtaining in adult animals. It may
be added that I purposely brought these observations to the
notice of the Society before the publication of Prof. E wart's paper,
since I treat the subject from a different standpoint.

In the first place, the callosities on both limbs of the Horse are
situated on the inner surface, whereas, if they represented
vestigial foot-pads, their position should be, i:)ri'mcii facie, on the
hinder aspect, as is the case with the ergot. It might, indeed,
be argued that they have changed their original position, but
of such a shifting there is no evidence in the adult. A second,
and perhaps more impoi'tant, objection to the foot-pad theoiy may
be drawn from the fact that the callosities in the fore-limb are
situated above the so-called knee-joint (carpus), and are therefore
altogether higher up than any of the foot-pads of plantigrade
mammals. Tiniest;, therefore, another shift of position has taken
place, the fore-callosities do not represent foot-pads. This argu-
ment, it may be mentioned, was used by Sir W. H. Flower to
disprove the theory that the callosities aie remnants of the lateral
digits.

The hind-callosities, on the contrary, are situated a short distance
below the joint of the hock (tarsus), and ai-e therefore on a part
of the limb, albeit on its inner side, which is included in the foot
of a plantigrade mammal. If, however, the front callosity be
regarded as serially homologous in a general sense with the hind
one — and this is an integral part of Prof. Ewart's theory, — it will
be evident that in the event of the former not being a foot-pad,
the same will hold good for the latter.

A third, and perhaps stronger, objection may be urged against
the foot-pad theory. On the assumption that the callosities of the
existing Eqiddce are vestiges of foot-pads, it is clear that these

1 See ' Nature,' vol. Ixvii, p. 239 (1903).

1903. J ON THE LIMBS OF THE EQUIB.E. 201

btructureb luuht have existed in the ancestors of that family ever
since the time when these ancestoi-s were plantigrade. But, so far
as I know, no ungulate was ever wholly plantigrade in lioth feet; the
nearest npproacli to this condition nlitaiuing in the Lower Eocene
Corj/phodoii, in which the hind-liml) was wholly plantigrade, while
the front one was partially digitigrade. It has thus to hea.ssuined,
on the foot-pad hvpothesis, that the front callosities of the Horse
have been fuuctioidess structures from a period antedating the
evolution of the Ungulata. 8uch a persistence, on exposed parts
of the body, of a wholly fuuctioidess structui-e seems very im-
probal)le, especially when the modifications are borne in mind
which, on this hypothesis, the horse-line must liave undergone
since the time when the callosities were functional structures.
Perhaps the case of the ergot may be cited against this argument ;
but it should be remembered that this structuie certainly acted
as a functional pad at a much later stage of evolution than could
possibly have been the case with the callosities.

Having now stated what appear strong objections, so far a.^; the
adult is concerned, against correlating the callosities of the Horse
with the foot-pads of polydactyle mannaals, it remains to consider
whether they can lie identified with any othei' structures. Those
familiar with the morphology of the Cervidce, will be awai'e that a
certain niunber of representatives of that family — notably the lein-
deer, the White-tailed Deer, the Mule- Deer, and, in a rudimentary
condition, the Elk, — are furnished on the inner side of the hock
with a glandular tuft corresponding very closely in situation with
the hind-callosity of the Horse. In fact, the only difference in the
position of the two structures is that the tarsal tuft of the Deer
in question is placed rather lower on the hock. From the fact of
its occurrence in Deer so widely separated fiom one another as
are the species mentioned, it seems evident that the tarsal gland
(which is doubtless a scent-organ) is a very ancient structure,
which was present in all the ancestors of the gi'oup, but has been
lost, probably from disease, in the great majority of Old World
forms.

Judging from their position, there would seem to be a certain
probability that the hind- callosities of the Horse and the tarsal
gland of the Deer are homologous structures.

With regard to the homology of the fore-callosity of the Eqv,id(T,
it may be mentioned that man}' Gazelles have tufts of hair
(" knee-brush&s ") at the knee (carpus), which are probably
glandular in origin. And it is possible (if the suggestion with
regard to the hind-callosities hold good) that these may represent
the fore-callosities of the Horse, for there seems no good reason
why the position of a gland should not have somewhat shifted in
two widely separated groups of mammals. Then, again, we have
the carpal bristles of cei'tain mammals, such as the Ooatis and
Dassies, already refei'red to as lieing regardeil by Mr. Beddard as
the remnants of a " scent-organ," — a. structure probably not far
removed in its nature from a gland. The occurrence of these

202 ON THE CALLOSITIES ON THE LIMBS OF THE EQUIDiE. [Mar. 3,

bristles in the Dassies (Procavia) is very important. Mr. Beddard
states that these are the only ungulates in which he has found
these bristles. Carpal callosities ai-e, however, described by
Dr. W. Leche^ as occurring in Wart-Hogs [Fhacochoerits); although
they are stated by their describer to be acquired, and not primitive
structvires. Whether the latter statement is calculated to modify
Mr. Beddard's opinion with regard to the nature of the carpal
bristles in the Dassies, I am, of course, unable to say. Of special
impoi'tance is the occui'rence of bristles in these structures, since,
even if hairs be found to exist on the callosities of foetal Eqiddoe,
this would be no bar to the svipposition of their glandular nature.

As regards the structure of the callosities themselves, it may be
noted that in the Horse both pads are of a distinctly warty nature,
and that the hind pair are certainly in a more decadent condition
than the other, being in fact on the verge of disappearing. In
the Zebras, on the other hand (in which the hind one has been
lost), the fore- callosity is larger and much less warty and also
situated higher up. In dried skins it is, in fact, much more like
the pale glandular patch of skin below the ear of a Reedbuck ^.
In this connection we have to bear in mind not only Mr. Beddard's
observations alluded to above, but likewise others by Mr. Bland
Sutton'^, in which it is pointed out that in certain Lemurs decadent
glands are actually converted into bunches of spines, which are
practically almost the same as warts ; that is to say, they are
hypertrophied growths of somewhat abnormal dermal tissue.
Hence there seems no primd facie reason why the callosities of
the Equidce should not be decadent glandular structures, the
decadence being more marked in the two pairs of callosities of the
Horse than in the single pair of the Asses and Zebras.

There is, however, another point which may have an important
bearing on the subject. Fiom the presence of a depression in
the skulls of Hipparion, Hippidium, &c., it is evident that
primitive Horses were furnished with face-glands comparable to
those of Deer ; such glands probably having a function somewhat
analogous to that of the scent -glands on the limbs of the latter.
If, then, the existing Equidce have got rid of their face-glands, as
being (perhaps on account of change of habit) useless, it is con-
ceivable that, for the same reason, they may have also discarded
their limb-glands.

If these suppositions (and they are but suppositions) be well
founded, it follows that a tarsal and a carpal gland must have
existed in the common ancestors of the Horses and Deer ; that is
to say, in the common stock of all modern Ungulates save the
Elephants and perhaps the Dassies. And it may be ui'ged that if
this were the case, traces of such glands ought to be met with in
Tapirs, Rhinoceroses, Pigs, Hippopotamus, &c. So far as I am

1 Biol. Centralblatt, vol. xxii. p. 79 (1902).

2 It would be important to examine the histological structure of ;the callosity in a
Zebra.

3 Proc. Zool. Sec. 1887, p. 369,

1903.] ON A FOSSIL OSTRICH FROM THE ISLAND OF SAM08. 203

aware, tlie only instances of structures wliicli can be regarded as
at all approaching this nature are tlie carpnl Ijristles of the
Dassies and the carpal callosities of Wa.rt-Hogs, the latter of whicli,
as already mentioned, ai-e considered to be of niodei-n origin. No
tiace of any structure compaiable with the hind-callosity of the
Hoise has, so far as I am awai-e, ever been detected on the tarsus
of any of the above-mentioned Ungulates.

If an objection of this nature be legai'ded as fatal to the gland-
theoiy (or sense-oi-gan- theory, foi' I i-egai'd the two as practically
identical) of the origin of the equine callosities, it will, I thiidc,
be still more so to the foot-pad hypothesis, since short-limbed and
polydactyle mammals ought to have I'etained traces of ancestral
foot-pads for a gi'eatei' period tlian long-limbed moiiodactyle forms
like the Kqukhe.

In conchision, I may state that it has been my object, not so
much to attempt to show what the equine callosities repi'esent, as
to demonstrate, from pahi?outological considerations, the impro-
bability of their being vestigial foot-jjads.

P.S. — I am informed that if a callosit}' be pared down, and a
finger moistened with the I'esnlting exudation held to a horse's
nose, the animal will follow nnywhei'e. If tins be true, it afibrds
strong testimony in favour of the gland-theory.

4. Note on some Remains o£ Strut /i to karat /leodoris Maj. of
the Island of Samos. By Rudolf Martin, of Basel
University.^

[Received Pebruaiy 11, 1903].

(Text-figures 30-34.)

In the Catalogue of the collection of Mr. Barbey at Valleyi'es
s./Orbe (Switzerland), published in 1894 l)y L)r. Forsyth Major'-,
besides a great number of mammals there is mentioned the femur
of a ratite bird, which, no doubt, belonged to the genus Slruthio
(Dr. Major could scarcely find any diflerence). Dr. Major,
considering the geological age of the deposits in whicli the bone
had been found (Upper Miocene), and recognising that it belonged
to a form dillerent fi'om the I'ecent Ostrich, created the navf species
Strathio karatheodoris.

Some time ago, Dr. Major received the fragment of a pelvis
from the Museum of the Vienna University (found in the same
place as the femur), which probably belonged to the same species,
and which is the subject of the following remai'ks.

I have to thank Dr. Major for having entrusted me with the
study of this pelvis and for having placed two photographs of the

' Coiniiuuiicutcd by Dr. C. I. Forsyth Major, F.Z.S.

- Le (u-^cineiit (issifi'iv do Mitylini et Cataloarue d'ossements fossiles ricueillis a
Mityliui etc., Lausanue, 1894.

204

MR. RUDOLF MARTIN OX A FOSSIL

pMar. 3,

femur (rostral and cavidal aspect) at my disposal, on which the
following description is based.

Evidently this femur (text-figs. 30, 31) belonged to a large
struthioiTS bird, as it differs little from the same bone of a modern
Ostrich. The size of the femur is somewhat larger than that of
the same bone of recent Ostriches at my disposal, but very probably
there Avould be no difli'erence between a, large specimen of Struthio
camehts and the fossil.

Text-fig. 30.

Text-fiff. 31.

Text-fig. 30. — Right femur of Struthio karatheodoris. Caudal view. 5: nat. size.
Text-fig. 31. — „ „ „ „ Rostial view. \ nat. size.

The following table gives the results of a compaiison between
the fossil in question and that of a medium-sized modern
ostiich : —

Length of Length of caput Smallest Width of

Species.

femur.

-(- coUum.

width.

trochlea.

Str. karatheodoris

36-0

10-5

5-5

12"4 mm,

Str. camehis

31-2

10-0

5-2

11-5 mm,

The measurements of the caput -f collum have been taken

1903.J

OSTRICH FROM THE ISLAXD OF SAMOS.

205

from the apex of the caput to tlie margin of the trochanter, and
that of the troclilea in the direction of its greatest expansion.'

A close stud}' of the bone, liowever, revealed some features in
which the fossil differs from the recent Ostrich, on the sup-
position that the photograph is not distorted.

In the fossil, the neck supporting the head of the femur is much
stouter than in the recent species and is much less constricted.
This is best seen in the form of the distal border of tlie collum.

^ Although nieasurements may include an error of, say, -L, they
give an idea of that feature, because the difference between the
two species exceeds that error. The only measurable line is the
plummet in the apex of the hea.l to tlie liiiext uspera, divitlin^'-
the rostral surface of the neck from that of the bone itself. °

Width of the neck = 1 .
StriitJi'io cameliis = A.

„ karatheodoris = ^

It seems to me that the groove in wiuCli the Hguuieiituin teres
is inserted is shallower in the fossil than in ,Struthlo cumelas. It
may be that the photograph gives a wrong impression, and I think
it would be better not to regard this feature as a specific one.

Text-fiff. 32.

"I' S/rnfh;,, k:,r.ith,'odoris. \ vmU'aX view
« & 6 = true -acral vertehi-.u. i.. ii., & iii. = first tin
,', iiat. size.

ee post sacral vertebra?.

206

MR. RUDOLF MARTIN ON A FOSSIL

[Ms

The trochanter and the li7ie(e asperce, of the proximal end of the
femur are not to be distinguished from the same parts in the
recent Osti'ich ; even the pneumatic foiumen in the caudal surface
is found in the same place.

The middle pai-t of the bone is just as in S. cainelibs, nd the
distal end diffei-s only in one particular : the fossa intercondy-
loidea is much elongated, so as to form a slight valley in the distal ^-
oi" I of the I'ostral suiface of the femur. In the caudal surface,
immediately above the proximal border of the trochlea, there is
also a funnel-shaped pneumatic foi'amen which enters the bone
obliquely (fi'om the mesial border of the bone, and is directed
towards the middle of the axis of the trochlea). As in the recent
Ostrich, there is a tuberosity" between the wide orifice of that
foramen and the mesial edge of the caudal surface.

The pelvis (text-figs. 32-34) difl^ers in some respects rather
considerably from that of the recent Ostrich ; but the individual

Text-fig. 33.

Pelvis of StrutMo karatheodoris. Dorsal view.

variation met with in the recent foim, and very piobably also in
the fossil, shows that the difliierence is much less than it seemed to
be at first sight, and very probably the extreme variations of
the two forms closely approach.

1903.] OSTRICH FROM THE ISLAND OF SAMOS. 207

The pelvis at my disjiosal is lepresented only by a fragment
from the last pre.9rtc;-rtZ to the t\\ivi\ postsacral vertebra (fi vertel)ra',
the first and the last incomplete), and in conneetion with it tlie
corresponding 2)arts of the ilia, *.e.the ncetabularaiid immediately
postacetabular i-egion. The acetahida themselves are not entii-ely
preserved, but only the caudal liorchn' of that on the left is quite
intact, and the antitrocltaiiter is broken away on both sides, so
that on the left side only its outlines can be deteiinined.

Text-fi^. 34.

Lateral view of the pelvis of Strufhio karatJieodoris. \ nat. size.

In the principal and most characteristic features, the fossil
does not difi'ei- from *S'. camelus, and there can be no doubt that
the ratite bii'd of Samos belonged to the same genus.

In the following featui'es the fossil pelvis agrees with oi- differs
from that of the recent Siruthio : —

i. The pelvis is compressed from both sides in the same mannei"

as in StrutJtio camelus (though tlie comi)ression is a little less

marked), so that the dorsal as})ect of it is much as in the

recent form.
ii. The sacral vertebra', of the fossil ai'e much stouter. The

same iri-egulai- sculptures occui', foi-ming lough longitudinal

ridges.
iii. The centra of the two " true sacral vertebra', " are, compai-ed

with the following centia, nari-ower, but better rounded

than in the recent Ostrich,
iv. The acetabulum has quite the same position as in aS'. camelus.
V. The plane between the venti'al border of the antitrochanter

and the dorsal edge of the foramen obturatorium is— in

comparison with >S. camelus — turned foi'ward, i.e. it slopes

more rapidly towards the acetabulum.
vi. The outlines of the antitrochanter ai'e just as in S. camelus,

and its size is also the same.
vii. The ilia of the fossil and the recent Ostiich are closely

similar. There is no difteience in the dorsal view of the

pelvis exceeding possible individual variation. The vertical
Proc. Zool. Soc— 1903, Vol. I. No. XIV. 14

208 MR. RUDOLF MARTIN ON A FOSSIL [Mar. 3,

distance between the antitrochanter and the lateral edge
of the area dorsalis is relatively longer in the fossil than
in most of the recent medium-sized Ostriches ; but on one
side this distance cannot exactly be determined in the fossil
(because the antitrochanter is broken away and the edge
of the area dorsalis is mu^ch rounded), and on the other side
the individual variation in this regard in S. camelus is very
great. The largest individuals of this species differ but
little from the fossil, and very probably the fossil pelvis at
my disposal must be referred to an adult and rather large
animal.

The distance between the dorso-mesial borders of the ilia is
somewhat shorter in the fossil than in most of the pelves of
aS'. camelus with which I have compared it. But it is in this respect
that individual variation has the greatest amplitude, and in the
largest pelvis of ^S'. camelus at my disposal this distance is equal
to that of the fossil.

/S. Icaratlieodoris. S. camelus.

Costal process of the 1st '' true sacral
vertebra" to the distal border of the
2nd postsacral vertebra 7*8 cm. 9*2 cm.

Width of the centra of the " true sacral

vertebrae" 2-5 1-8

Width of the centrum of the 2nd postsacral

vertebra 2-8 2-3

Greatest width of the dorsal surface of the

pelvis 12 12-2

Plummet in the dorsal border of the anti-
trochanter to the median plane ca. 1 2 12

Plummet in the dorsal border of the anti-
trochanter to the plane of the dorsal
surface of the pelvis ca.6'5 4

The results show that there are great differences in the pro-
portions of the pelvis of the two species which justify a
separation of the fossil from the recent Ostrich. The individual
variation, however, does not allow us to place the two forms in
different genera, because it greatly reduces the differences above
mentioned.

I have said that the separation of the two Ostriches in different
species was justified. On the other hand, the question arises,
whether or no the differences between the two forms be not the
consequences of a pressure acting during the great tectonic trans-
formations in the region between Asia Minor and Greece. A
study of the other fossils of Samos, however, gives no evidence of
such a force ; there is only a lateral compression to be seen, a result
of the pressure of the weight of the more recent strata.

These considerations add to the importance of the features which
I have selected for special comment, Avhilst on the othei' hand other

1903.] OSTRICH FROM TOE ISLAND OF SAMOS. 209

featm^es lose their systematic value. But I have sliown that the
latter had really already lost their importance because of their
individual variation.

The species Struthio karaiheodoris Maj. is therefoi'c to he kept
separate.

What are the relations between ^S'. karatheodoris and ^S'. asiaticus
from the Siwalik Hills? I can now give further information
on *S'. asiaticus. A comparison of the fossils with the figures on
the unpublished plate R of Falconer's ' Fauna Anti(|ua Sivalunsis,'
and those accompanying Davies' ^ paper and Lydekker's 'description,
did not reveal any differences, except that the drawings in
Falconer's Atlas are much better than the others. Davies found
that the principal difference between S. asiaticus and >S'. camelus lies
in the greater stoutness of the cervical vertebra) in the former ;
and Lydekker kept the two species sepai\ate onlj^ on account of
this feature. The answer to the question, whether the greater
stoutness of the sacral vertebrae on one side and tliat of the cei'vical
vertebraj on the other indicates a special i-elationship, cannot be
given now. There is no evidence for such relations ; and I, for
my part, eonsidei- this character in the fossil forms (compared with
the modern Ostriches) only as more primitive.

There may be i-eason to unite the two forms in one species ; but
considering their different geological ages', I think it will be better
not to do so. But there can be no doubt that Struthio asiaticus
is in direct relation with >S'. karatheodoris, i. e. that the former is
a descendant of the latter, as Dr. Major' supposes the whole
Siwalik fauna to be a later and transformed generation of the upper
Miocene fauna of Pikermi and Samos.

^ It is noteworthy therefore that S. karatheodoris and S. asiaticus
give us no evidence for a specialisation of the Struthionidte in
Southern Eurasia", and a consequent emigration into Southern
Europe, S}a-ia, and Africa, but support rather the view that the
order of migration took place in the opposite direction. However,
the genesis of the Struthionida; cannot, as Burckhardt ' is inclined'
to suppose, be associated with the Miillerornithida- ; for the
geological age of the former is opposed to such an hypothesis.

By the discovery of a Struthio in the island of Samos, the eo-o-
on which the species of Struthio chersonensis Brandt ' has been
based is of some interest. Though the circumstances surrounding
its disco\'ery may be somewhat obscui-e, yet the size and proportions
show that it cannot be the egg of a modern Ostrich ; tliat it really

] Davies, Geol. Mag-. 1880. ^ Lydekker, in Palaiontologiu Inclica, 1881-8G

3 Lydekker, Fossil Vertebrata of India : Uecords Geol. Survey, India, vol. xx. 1887
^ Op. eit. C;onii)t(!s rendus des SiSancos de I'Acad. d. Sci., Oct. 18HK. (An account of
the fauna of Samos.) Atti della Societa Toscana di Scienze Natural!, I'roc -verb'di
vol. V. 3 July, 1887, p. 272. (Letter addressed to Trof. Menegliini, in which'
Dr. Major shows the difference between the mammalian remains of Samos and th o
of Kos.)
i Burckhardt, " Uber Aepyornis," Pal. Ahh., Jena, 1893, p. 21.
'- Burckhardt, " Das Problem des antarktischen SchopfunMcentrums " &c Zool
Jahrbiicher. 1902, p. 2(1. ' '

' Bull. Ac. Imp. Sc. St. Petersboura-, vol. xviii. 187.^.

14*

210 MR. F. E. BEDDARD ON [Mar. 3,

is an egg of a Struthio has been proved by von Nathusius. Its
geological age cannot certainly be determined because it was found
floating in a river. Brandt supposes that it had been embedded
in the bottom of the river and had been worked out by the water.
The age of the sti'ata forming the bed of that river is supposed to
be that of the strata in which von ISTordmann ^ discovered the
mammalian remains in S. Russia (environs of Odessa), and therefore
that of the breccia of Pikermi ; but it may be much younger,
because Nordmann^ does not separate the Tertiary from the
Pleistocene. In the paper above mentioned, he attributes the
bone-beds of S. Russia to the "offenen Diluvium," and says that it
has the same geological age as the bone-beds of the Val d'Ai-no.
Therefore we have no evidence of the exact geological time during
which a Struthio lived in South Russia, and the egg is in con-
sequence of little importance in regai-d to this question. Moreover,
it seems to me more than doubtful to assign this egg to the
modern Ostrich, and "very likely to the species occurring at
Samos," as is done in Nicholson and Lydekker's ' Manual of
Palaeontology ' (p. 1228).

5. On a new Genus and two new Species o£ Earthworms o£
the Family Eudrilidce, with some Notes upon other
African 01igocha?ta. By Frank E. Beddard, M.A.,
F.R.S., F.Z.S.

[Received March 3, 1903.]

(Text-figures 35-38.)

The first set of specimens referred to in the following descriptions
form a part of the collection in the British Museum ; I am greatly
indebted to the kindness of the Director for allowing me the
opportunity of examining them. They belong to two species,
both of which are undescribed. The first is a third species of the
genus Stuhlmcmnia, which I call

Stuhlmannia michaelseni, n. sp.

I have examined so many individuals of this form of Stuhl-
mannia, and the agreement between them is so close, that I have no
hesitation in regarding it as a distinct species, which I name
after the founder of the genus. Dr. Michaelsen ^.

The worms were collected by Mr. S. L. Hinde in the Mt. Kenya
district. There were thirty mature examples.

The general appearance and proportions of this new species are
quite the same as in ^S*. variabilis. The length reaches 100 mm.
and the diameter 2-2*5 mm. The colour is a yellowish grey ; the
clitellum is greyer.

1 Nordmann, in ' Jubilteum semisaeculare Fisclieri de Waldheim,' Moscou, 1847.

2 Nordmann, Palaontologie Siid-Russlands, 1858.

3 JB. Hamb. wiss. Anst. vii. (1890), p. 24.

1903.]

NEW SPECIES OF EARTHWORMS.

211

Tlie setye ai'e strictly paired and the ventral pair of segment xvii.
appear to be missing. The clitellum occupies segments xiv.-xvii.

The median orifices of the spermatheca and of the spermiducal
glands are very obvious, and frequently lie upon projecting papillae.
The spermathecal poi'e opens far back upon the xiiith segment.
The male pore opens far back upon the xviith segment, or even
intersegmentally (xvii./xviii.). The penial settc can often be seen
to protrude from it. Among the external characters, however,
there is one which distinguishes the present species fi'om its allies,
and that is the total absence of the penial pi'ocessof the body- wall.
There was not the least ti'ace of this oi-gan obsei'vable in any of
the individuals. It is true that this penis is not always to be
seen in Stuhlmanida varioMlis ; but I do not think that it would
be likely to be absent fi-om thii-ty individuals selected at i-andom ;
and this absence is therefore one of the reasons which lead me to
create a new species.

Text-fiff. 35.

Dissection of Stulilmannia michaclseni.

xi, xii, segments flevi'u and twelve ; L.V., intcstino-tegumentav.v trunks ; o.d., oviducal
apparatus ; Sp., spermathecal sac ; Sp.ffl., spermiducal glands.

The accompanying draAving (text-fig. 35) illustrates a general

212 MR. p. E. BEDDARD ON [Mar. 3,

view of the viscera seen on cutting open the body-wall in the
dorsal median line. No such figure of the anatomy of this genus
has been published up to the present ; and it is convenient to
show the relations of the different organs and their comparative
sizes.

There are points in the anatomy of the genus Stuhlmannia
which are illustrated by that drawing, and to which attention does
not appear to have been called. The last pair- of hearts lies in the
eleventh segment, as in so many other Eudrilids, for instance in
the genus Polytoreutus ; and the septum which should separate
segments xiii. and xiv. is nearly missing. I am disposed to
associate this latter fact with the presence of the ccelomic sacs
svirrounding the gut, which may have been developed at the
expense of the septum, as in Euch-ilus ', in which genus the
corresponding septum is also much reduced. Another matter to
which I would wish to direct attention, is the existence of intestino-
tegumentary trunks, which have not yet been recorded in this
genus. They are, of course, of wide occurrence among earth-
worms.

A further point of some little interest is visible in the sketch
exhibited herewith. In my ' Monograph of the Oligochteta,' '" I
pointed out the existence of at least occasional asymmetry of the
female reproductive organs shown in the presence of only a single
receptaculum ovorum, that of the opposite side of the body being
absent. In other specimens ^ I found precisely the same state of
affairs. The species that I examined was, I believe, Stuhlmannia
va7'iabilis.

In the present species of StiMmannia there is exactly the same
asymmetry, the I'eceptaculum being only present upon one side,
and that the right. Or, to be more accurate, the receptaculum is
possibly present also on the left side, but is quite rudimentary,
and, I imagine, f unctionless. The " Eitrichterblase," as Mithaelsen
has termed it, is present on the left side, and is simply a loop of the
oviduct, the two sections of the tube running in close contact side
by side ; just opposite to it is a small spherical projection. A
lumen is present in this, which is the rudimentary funnel, but
the lumen is no wider than that of the oviduct elsewhere, and
there is no question of any free communication with the body-
cavity outside ; thei-e was no break to be detected in the muscular
wall of this projection, which perhaps should be regai'ded as funnel
+ receptaculum.

A series of sections through the rudimentary funnel and the
adjoining parts of the oviduct confirms the appearances displayed
in a preparation mounted in glycerine. The rudimentary funnel
(or funnel + receptaculum) is but a slight protuberance, which is
traversed up to its very end by a blindly ending branch of the

1 Beddard, " On the Gonad Ducts and Nephridia of Eudrilus," P. Z. S. 1902,
vol. ii. p. 89.

2 Oxford, 1895, p. 602.

3 " On some Earthworms from British East Africa, &c.," P. Z. S. 1901, vol. i.
p. 355.

1903.] NEW SPECIES OF EARTHWORMS. 213

descending limb of the oviduc.il loop, which constitutes the
" Eitvichterblase " of Michuelsen. No communication between
this slioi-t diverticulum of the oviduct and the general crelomic
cavity could be detected. Evidently, however, it is the forerunner
of the funnel which opens into the fully developed receptaculum,
or, possibly, of both. It is noteworthy that the oviduct
is divisible into two regions. The section lying nearest to the
spermatlu'cal sac is of a wider calibre than that lying nearer to the
external orifice. Furthermoi'e, the lumen of this wider section
is a straight tu])e, while the lumen of the narrower section of
the oviduct is sinuous in its coui'se within the muscular sheath.
It is possible that the wide straight section is really to be looked
upon as a greatly drawn-out funnel, drawn out by rea,son of the
gi-owth of the spei'mathecal sac.

The walls of the oviduct, as is the case with other Eudrilida?,
are very stout and muscular ; and attention has ah'eady been
called to the fact that part of tlie oviducal tube runs a, sinuous
course, independent, therefoi'e, of its musculai- covering. These
facts, coupled with the conditions obta,ining in EiulrilK.s ', where
the oviduct runs for a considerable space actually within the
septum, lead me to consider that the muscular wall of the oviduct
in StiMmannia may be a, purely adventitious slieath, not belonging
to the oviduct at all, but consisting chiefly of tlie otherwise
missing septum between segments xiii./xiv. Its continuity with
the receptaculum can at any rate be explained on this view, which
is in any case not at variance with the other facts to which I have
just called attention.

The present species, in correlation, it is to be presumed, with
the entire absence of a penial process, does not possess the median
unpaired bui'sa pi'opulsoi'ia which chai'acteiizes *S'. variabilis. In
the former point ^^'^''^^iWy, but not certainly, it agrees with
S. gracilis. Di-. Michaelsen states, only as a possibility, since he
only examined two specimens, the absence of a copulatoiy process
in S. gracilis ; furthermore, a thickening of the integument of
segment xv. appeared to him to be possibly intelligible as a not
fidly developed penis. With I'egard to the presence or absence of
a Inu'sa propulsoria he is silent. Stuhhmmnia michaelseni, however,
caiuiot be confused with S. gracilis by reason of the characters
of the penial setag. They ai-e provided at the end with a row of
shaip denticles on either side as in aS'. variabilis. In short, I
think that a case has been made out for the creation of a new
species of this genus StiMmannia.

Bettonia lagariensis, n. g, k sp.

Of this apparently new genus and species I have examined but
a single specimen, complete and sexually mature. It measures
90 mm. by 4-5 mm. in breadth. The colour is bluish violet. It

1 Bcddavd, " On tlie Gonad Ducts and Nqihiidia of Etidrilus," P. Z. 8. 1U02,
vol. ii. p. 89.

214 MR. F. E. BEDDAEB ON [Mar. 3,

was collected by Mr. Stuart Betton, at Lagari, British Central
Africa.

The prostomium is continued by grooves over about half of the
buccal segment.

The setee are wider apart in the case of the ventral couples than
in the case of the lateral.

The limits of the clitellum are a little obscure ; it appears to
embrace segments xiv.-xvii. inclusive.

The chief, indeed practically the only, reason which leads me
to separate this Eudrilid generically, is the condition of the repro-
ductive apertures. The rule in the family is that the apertures
of the sperm-dvict and of the spermathecse are unpaired and
median in position.

Text-fig. 36.

V i

Ventral surface of Bettonia lagariensis, ?, sperma thecal pores; <?, male pore.

There are, however, a few exceptions, such as Eudrilus itself,
and an apparently close ally of the present genus, viz. Eriiinoscolex.
Bettonia offers a third arrangement of the reproductive pores.

1903.]

NEW SPECIES OF EARTHWORMS.

215

The spermathecal pores ai-e paiied, while the male poie is single
and median (text-fig. 36, p. 214).

The two former poies are upon the boundary line of segments
xii./xiii. ; each is somewhat eye-sliaped in contour and corr&sponds
exactly in position to the outermost of the vential couple of setae.

Tlie single and median male pore lies on the boundaiy line of
segments xvii./xviii. Jt is very large and conspicuous, with
radially folded maigins, indicating, pei'haps, the possibility of the
extrusion of the bursji propulsoria.

The internal oi-gaiis, unfoi-tunately, were much softened, and
therefore the female reproductive apparatus was rather difficult to
deciphei'. The nature and i-elations of the other organs of the
body were not so diflicult to asceitain.

As in many, if not in all, Eudiilidse, the last pair of hearts
occurs in segment xi. In front of these were four distinct pairs.
The doi'sal vessel is single thi'oughout. I have not been able to
study the alimentary system in detail, but I have ascei-tained
that the present species is a member of the section Eudrilacea l)y
virtue of the existence of an unpnired ventral median alimentary
gland in segment x., which showed on a niicroscopical examination
the usual laminate structure of these glands, not at all to be
confounded, even in a badly preserved specimen like the present,
with the totally difterent though corresponding organs in the
other subfamily of the Eudrilidae. There are also a pair of these
glands in segment xiii., of an oval form, which contain abundant
crystals, such as are met with in other genera.

Text-fig. 37.

Tei-mination of the male efferent apparatus of Bettonia lagariensis.
B.p., Bursa propulsoria ; Sp.ffl., spermiducal glands.

The only conspicuous parts of the male generative system are

216 MR. F. E. BEDDARD ON [Mar. 3,

the sperm-sacs and the terminal apparatus, which opens on to the
exterior in the xviith segment. The sperm-sacs are in segments
xi. and xii., and occupy a considei-able space in those segments.
They are simple, solid, sac-shaped structures, not racemose in form.

The spermiducal glands are paired. Each of them (see text-fig.
37, p. 215) is shortish and rather thick, tapering somewhat towards
the tip. The last third of the gland is bent forward and lies above,
parallel to and in contact with the anterior section of the gland.
This arrangement occurred on both sides of the body. Each
gland is sharply constricted at its opening into a large median
bursa propulsoria lying below the ventral nerve-cord. This latter
sac, however, presents obvious signs of having been produced by
a fusion of two sacs ; for posteriorly it is completely double. It
is into each of these posterior lobes that the spermiducal glands
open.

My description of the female appaiutus must unfoi'tunately be
incomplete. The oi'gans, as already stated, are paired. The
spermathecfe, near to where they open, have very thick musculai'
walls ; and this i-egion at least is enveloped in a coelomic sac, as is
the base of the spermatheca in the species of Pareudrilus (?), with
which I deal later in the present communication (see below).
How this ccelomic sac is otherwise related to the spermatheca
and to the receptaculum ovorum I am unable to state. The latter
organ presents no noteworthy peculiarities, and the ovidiict leads
from it to the exterior, on the fourteenth segment.

On a Sjyecies of Pareudrilus.

I believe that a number of individuals belonging to this genus,
which were collected by Mr. Crossland, may represent a new
species. But I am unable to speak with absolute certainty on the
matter, since the material was not in good order for investigation,
and since the specimens of P. papillata examined by Michaelsen ^
were likewise much softened by evaporation of the alcohol ; if the
worms upon which I report here are not referable to P. 2)apiUata,
then the species is unqviestionably new.

But whether the species be new or not, I have something to add
to what is known about the structure of this genus.

The dimensions of my specimens agree apparently with those
given by Michaelsen for his sj)ecies. The length was some 100 mm.
and the diameter about 3 mm. The dark purple colour, turning
to yellow below and in the clitellar region, is like that of my
Pareudrilus stagncdis. The setee are closely paired, and I observed
a tendency in the neighbourhood of the genital pores for one seta
of a pair to be lost or not developed. I do not refer to a mere
dropping out ; on examining the cuticle, occasionally no pore was
to be noticed in the place where such a pore (through which the
seta is extruded) should be. The irregularity of this state of afi'airs

' " Die Regenwurmer Ost-Afrikas," in Deutsch-Ost-Afrika, p. 11.

1903.] XEW SPECIES OF EARTHWORMS. 217

does not incline me to describe in detail the cases observed, which
were, moreovei-, not many. I can eoniirm Michaelsen's statement
that there are no genital or any specially modified setfe in any
region of the body.

In the external sexual chai-acteis I find some little diflerences
from the description of Michaelsen, and it is pai-tly on this account
that I am dis[)osed to regard my specimens as belonging to a new
species. The clitellum is perhaps a little more extensive, xiii.-
xviii. or even xix. ; but that is a slight difference. I find that the
pores of the paired spermathecie are situated definitely on the
l)order-line of segments xiv./xv. and not upon xiv. They are,
however, in the line of the ventral couple of setre. On the other
hand, I agree with Michaelsen in placing the pores of the spermi-
ducal glands upon segment xvii. just in front of the ventral setae
of that segment. As to the location of the genital papillae, l_>r.
Michaelsen and I have not found the same conditions in the
specimens studied. In my specimens, the constant rule appears to
be the pi-esence of two pairs of pajiilLe of which the most anterior
are on segment xvii., just behind the ventral seta^ but on a line
with the outermost setre of the couple ; on the following segment,
the xviith, there is a precisely similar paii' also behind the ventral
set;e, but in this case corresponding to the innermost seta of the
couple.

The papilhe aie faiily conspicuous, and aie to be noted upon the
cuticle when stripped off. The area is slightly laised and is
studded with the mouths of large glands. Corresponding to the
papillre inteinally, are glands to which I shall recur in describing
the anatomy. These papillae were nevei- absent.

I could discover no nephridial pores ; but I imagine that this
species, like Fareudrihcs stagualis and some other Pareudrilacea,
will turn out to possess a ramified system of nephridial end-
tubes in the integiunent. In any case the nephridia, viewed
inteinally, appear to be paired structures. As to the alimentary
tract, I find a gizzard far foi-ward, in the sixth or seventh segment,
and I have not been able to find calciferous glands. The last
heart is in xi. The funnels of the sperm -ducts are in x., xi. The
two paii's of sperm-sacs, in xi., xii., are tongue-shaped.

As to the female generative system, my observations do not
altogether agree with those of Michaelsen ; but it is no disparage-
ment to that excellent and accurate obsei'ver to suggest that the
condition of his specimens may possibly have led him into some
slight error. Otherwise I must place my specimens not only in a
new species, but in a new genus. The ovary I have not been able
to find at all. This in an Eudrilid is not of course surprising;
for as a i-ule the ovary disapjieai'S early, having transfei-red its cells
to the egg-sac. On the other hand, while it is unlikely that I
could have missed so conspicuous an object as the large o\'arian
sac figured by Michaelsen, it is equally unlikely that he wa,s misled
by bulging septa or other structures into stating the presence of
such a sac communicating with the egg-sac. It seems to me

218 MR. F. E. BEDDARD ON [Mar. 3,

therefore to be possible that, after all, in spite of undoubted
resemblances, Michaelsen and I have examined difterent species
and indeed different genera.

The female organs of the species which I have examined show
a nvxmber of not uninteresting features. The two most salient
parts of that system, which are visible on a dissection, are the two
spermathecse and the egg-sacs. The spermathecse are sausage-
shaped, and distinctly divisible into two regions. The proximal
part, i. e. that nearest to the external orifice, is strongly muscular,
and indeed is enveloped with stronger muscular bands than is the
bursa propulsoria of the spermiducal gland. Its epithelium is
perfectly continuous with the general epidermis of the body- wall,
and it has every appearance of being formed as an ingrowth from
the exterior. The distal region of the spermatheca has very thin
muscular walls, much thinner than the walls of the glandular part
of the spermiducal gland.

The interior has an epithelium which is raised into folds. I
cannot speak of the histological characters of the cells, as the
material was not sufficiently good. There is a very close resem-
blance, on a superficial view, between this spermatheca and the
spermiducal glands. Indeed, on a cursory inspection, they
might be taken for consecutive pairs of either spermathecse or
spermiducal glands. The next most obvious part of the female
reproductive system is a very large mushroom-shaped body closely
adherent to the septum dividing segments xiii./xiv. This body is
stalked, and appeai'ed, on dissection, too large to be identified
with a receptaculum ovorum (or egg- sac). Nevertheless it is the
egg-sac, and by virtue of its large size it appears to be precisely
like the egg-sac of P. pajnllata described by Michaelsen. On a
closer inspection, a fine tube, apparently leading from the stalk
of the egg-sac to the muscular part of the spermatheca, was
apparent ; this seems to correspond to the narrow tube (sg.) figured
by Michaelsen.

A series of longitudinal sections through the body showed more
accurately the relations of these diverse organs to each other. I
find that the spermatheca is entirely independent of the rest of
the female apparatus, and that its cavity does not communicate
with the narrow tube arising from the egg-sac. That narrow tube
exists, as I have already mentioned ; but on reaching the base of
the spermatheca, i. e. the muscular end portion, it dilates into a sac
which entirely surrounds the muscular part of the spermatheca,
but does not, so far as I could ascertain, open into it anywhere.
The conditions, therefore, are those of such a genus as Hyperio-
drilus or Heliodrilus, where a true spermatheca is invested by a
ccelomic sac. Now, though the difference may appear to be slight,
I am disposed to think that it is important, and that a sperma-
theca which has no communication with the egg-conducting
apparatus is essentially different from a spermatheca which has
such a communication. It seems to me, for example, to be wrong to
compare the spermathecal sac of Lybioclrilus or Stuhhnannia with

1903, J NEW SPECIES OF EARTHWORMS. 219

the spermatheca of JTeliodrllus. For this i-eason I cannot agree
to Mieluielseu's ])laciiig- of my genus Alvaiiia within tlie genns
IL/pertndrilus. Tlie former Iims a, true spermatheca, homologous
witli tliat of otlier non-Euih'ilid eartliworms, while the latter hi\«
not. This also is the case with the genus or genera with which we
are dealing n-ow. Pm'eiulrilus staynalis has a spermathecal sac
which communicates with the c(elomic sacs envolving the ovary;
" Pareudrilas " jmpillata has not. It is possible, therefore, that we
should revive Michaelsen's Unyoria for the latter species. The
stalk of the egg-sac, as might be expected, lodges the funnel of the
o\iduct, or, to speak more accurately, the greater part of the funnel.
The exact conditions obtaining ai'e the following. In a series of
sections it may be seen that the nai'row tube communicating with
the ccelomic sac siu'i'ounding the proximal end of the spei-matheca
runs forwards and opens into the caxity of the xiiith segment by a
wide oiitice ; its walls are therefore continuous with, and no doubt
developed from, the septum bounding segment xiii. posterioi-ly.
The uppei- " lip " of the ostium is covered by the cubical cells of
the oviduct, which hei-e opens freely into the cavity of the xiiith
segment. Further on in the series of sections, the mouth of the sac
surrounding the spermatheca is closed, and the tube oj^ens into the
egg-sac through a wide tube which is entii'ely lined by the cells of
the oviducal funnel ; these are, of course, perfectly continuous with
those cells wdiich lie in the xiiith segment. The part of the
oviducal funnel which lies in the xivth segment appeals to be
divided into two, and to open by as many mouths into the hu"-e
egg-sac, which is so kidney-shaped as to be nearl}^ divided into two
sacs; I cannot pretend to an accurate description of the funnel and
its various foldings. It is clear, however, that the conditions which
obtain ai'c those of the more typical Earthworms, where the funnel
opens partly freely into the xiiith segment and is pai'tl}^ I'eflecterl
so as to open within the egg-sac. Now, the simpler forms of
Eudrilida?, such as the genus Eiidriloides, are distinguished by the
fact that the ovary is unenclosed in any sac, and that the funnel
of the oviduct opens piecisely as has been just described in Pareii-
drilus jyapillcda. On the othei' hand, in the more complicated
foi-ms, such as Stuhhnaniiia, the sacs containing the ovaries
envelop also the oviduct-funnel and communicate with the
spermathecal sac. The species which forms the subject of the
present remarks is plainly intermediate between these two
extremes ; and for that reason, as I think, deserves geneiic
sepaiation fiom Pareridrilus.

§ On the Sjiermatojihore.
The spermatophores of the Eudrilida^ have not been much
studied, and, so far as I am aware, are known only in the o-enera
Stuhlmannia and Poly tor euttos, in which I ha^-e m}'self described
them. Many, if not most, of the individuals of the present .species
which I examined had a single spermatophore in both the
spermathecae.

220

MR. F, E. BEDDARD ON

[Mar. 3,

As willTje seen from the accompanying drawing (text-fig. 38),
the spermatophore consists of a globular swelling followed by a
long thin tvxbe. The sperm, which appears blackish in glycerine
pi-eparations, is limited to the spherical or nearly spherical recep-
tacle at the distal end of the case. The spermatophore, as will be

Text-fio-. 38.

Spermatophore of Paretidrilns sp.

also apparent from the figure referred to, is of an elegant form,
not precisely corresponding to that of the spermatheca in which
it lies. The swollen and globular receptacle of the sperm is not
more than one-third of the length of the " stalk," which reaches
down to the very mouth of the spermatheca. At its termination the
Avails of the spermatophore project in a ring-like fashion ; a state of
afi'airs exactly recalling the spermatophores of the Tubificidfe, and
of the genus Stuhlmannia among the Eudrilidse. No doubt the
shape is due to a moulding upon the walls of the spermatheca ;
but the state of preservation of the specimens does not enable me
to give details. The walls of the spermatophore appear to be firm
and thick, and rather brittle in consistency. In teased prepara-
tions the rupture of the stalk was invariably a clean fracture.
The walls aie fibrous in appearance, and of the usual pale brown
colour that is generally associated with chitinous membranes. The
extremity of the tube, i. e. that which is nearest to the mouth of
the spermatheca, is open ; the other end is quite blind. The

1903.] NEW SPECIES OF EARTHWORMS. 221

shape of the whole sperniatophove is sufficiently ("hicidated by the
drawing icfefied to.

§ Note, oil, the CUlallniit (^'Alma stuidmainii, dial on a possibti/
new species of the genus Alma.

I believe that a note by myself ' upon the clitellum and
spermatophores of a West- African species of the genus Ahna is
the first recoi'd of the extent of the clitelhun in that genus. Like
other aquatic forms, Alma seems to be characteiised by a sea.sonal
development of the clitellum ; and hitherto, with the exception
just mentioned, no one appears to have seen or at least described
this oigan in that Geoscolecid. In the species to which 1 have
just referred, the clitellum was found to extend from segment xlv.
to Ixxxv., a position which is quite unlike that found in any other
Geoscolecid and present only in a few Lumbricids. This is an
additional reason for associating the genus Alma more particularl}-
with Criodrilus, as is done by Michaelsen ; for both these genera,
though I'eferable to the family Geoscolecidne, have many points of
kinship to the Lumbricida\ As, howevei', up to the present time,
but one species of the genus Ali/ia has been descril)ed in the fully
mature condition, it is possible that the position and extent of the
clitellum characterising that species are not normal but exceptional
in the genus. Theiefoi-e I do not hesitate to desciibe the con-
ditions occurring in a second species of the genus, which I owe to
the kindness of Mi-. Cyril Orossland, who collected specimens on the
shore of Victoria ]Sr3'anza, among weeds cast up by the waves. 1
have tw^o fully matuxe examples of a species which I believe to be
identical with Dr. Michaelsen's Alma stuhlmanni. The dimensions,
howevei-, are rathei' less ; only one of the two examples was quite
intact — the other had lost the hinder end of the body; in the
complete example, measurements showed a length of 120 mm.
The other ujight have l)een slightly longer, as it was rather thicker.
In both, the penial appendages were i-ather longer than those of the
original specimens described by Michaelsen. I found them to be
10 millimetres long ; in one example the two weie unequal in
size, one penial appendage only mesisured 6 mm. The structure
of these appendages is usually chaiucteristic of the species. The
worms which T ha\e examined agreed in almost eveiy detail with
the description of A. stuhlmanni as given by Michaelsen'-.

I n»ay i-emark, howevei", that thei-e were only two setse at the
free end of the penial process, and, indeed, one of these had
dropped out. The two papilkv upon which these setje ai-e placed
were paiily encircli^l by a, horseshoe-shaped region of specially
tdandular epideiiuis, which was conspicuously marked out from
the rest of the int(>gument covering that process. It is conceivably
this region which secretes the spermatojihoi-e. In addition to this,

1 "On theClitolluni ami S))iTiiiiitoi)limi'-^ utaii Aiiiidiil of tlic Gi'iius^/iwrt." Proc.
Zool. Soc. 1901, vol. i. p. 21ii.

- " Die Regeiiwiinner Ost-Afrikas. ' in Dinitseli-Ost-AtViku, 181)G, ]). k

222 PROF. NEWTON ON THE WHITE RHINOCEROS. [Mar. 17,

I fancy that the single papilla at the base of the penial process is
rather nearer to the base than is figured by Michaelsen, and much
nearer than in the West- African form, and, moreover, it appeared
to me to be not symmetrical in its position. The clitellum was
not seen by Michaelsen at all. In my specimens the clitellum
occupied the same segments ; but, as in the species described by
myself from Western Africa, that geneiative region was somewhat
undefined in its beginning and ending. The greatest number of
segments referable to the clitellum in the West- African species
are from xlv.-lxxxv. ; but the clitellum was only fully developed
upon segments xlvii.-lxxxii. Its general appearance was precisely
like that of the species which I have just mentioned. In the
present species — and the observations apply to more than one
specimen, and are therefore all the more reliable as an expression
of normal conditions — the clitellum was much shorter, only
extending from segments xl.-lxxi. This rather leads me to the
inference that the species with which I am concerned here is in
reality difierent from that which I described fi'om McCarthy Island
on the Gambia, and referred to A . stuhlmanni. The present worm
is undoubtedly A. stuhlmanni; and it seems to be necessary, on
account of the difference in the clitellum, to use another name for
the West- African form. I would propose, thei-efore, to call the
latter A. hudgetti. It is, however, clearly a very close ally of
A. stuhlmanni. It is interesting to note that the species of this
genus go more or less in couples. The East- African A. nilotica
corresponds to my A . millsoni from West Africa, while A . stuhlmanni
seems to be neaiest to the form which I propose to name J^. hudgetti.
At present more information is wanted about A . emini ; but it
appears to be formed rather after the plan of A. millsoni and
A . nilotica. For in those species there are special setse on the penial
processes, while in A . stuhlmanni and S. hudgetti there are setse of
the same pattern as those on the body generally.

March 17, 1903.

G. A. BouLENGER, Esq., F.E.S., Vice-President,
in the Chair.

The Secretary exhibited, on behalf of Prof. Newton, F.K.S.,
three photographs of the White Rhinoceros {Rhinoceros simus),
sent to him with the following letter by Mr. C. R. Saunders,
C.M.G., Chief Magistrate and Civil Commissioner inZululand: —

Eshowe, Zululand,
^ „ 6th January, 1903.

Dear Sir,

I received a letter from you in August 1900, following
on an account, written by me in ' The Field,' of an interview I had
with White Rhinoceroses about that time. I did not answer
your letter at the time, hoping I should be able before long to
send you a photograph of the living animal. This, however, I

1903.]

PROF. XEWTON OX THE WHITE RllIXOCEROS.

223

liave not yet been iible to procure, although I visited the reserve
in which they lived for this special purpose last winter. Their
ti-aces were abundant, but my time was limited, and they could
not be found. Thei-e are, I believe, about ten of these animals
living in that reserve, and I do not despair of yet obtaining
a iihotograph of them in life, in wdiich case I shall be pleased
to send you a copy of it.

Early last December two of the nninuds (lK)t]i ))ulls), one a very
old one and the other not full-grown, str;iyed out of the reserve
into one of the native locations and were killed. I obtained
three photogriiphs of one of them, the old bull, t;iken l)y an
amateur two or three days after it had died. I am forwarding by
the same mail as this a. copy of each of these photographs, wliich
you are welcome to, and which, 1 think, demonstrate the fact that
they were taken from a specimen of the While Ithinoceros, although
the carcass was a good deal distended.

The killing of these two Rhinoceroses was most unfortunate.
They suddenly appeared among some native kraals, and the men
went out and attacked them with sjiears. The young one was
killed outright; that of which the photograph was taken travelled
a long distance after being wounded, aii<l was not found mitil
some days had elapsed. Yours faithfidly,

Alfred i\eivion, Esq. C. R. Saunders

As tliese photogi'aphs weie piobably the only representations
Text-%. 39.

Kircnlly-killtd liliiiiorinis siiinis, aAn^i $. Dec. 1902.

Proc. Zool. 8oc. — 1903, Vol. 1. No. XV. 15

224 MR. OLDFIELD THOMAS ON A NEW MONKEY. [Mar. 17,

ever taken of this animal in the flesh, it seemed well worth while
reproducing two of them (text-figs. 39, 40),

Text-fig. 40.

Recently-killed Rhinoceros' simus, adult $ . Dec. 1902.

Mr. Oldfield Thomas exhibited the skin of a Chinese Monkey,
which had been obtained from a hunter by Mr. Henry Brelich,
and presented by him to the National Museum. It appeared to
i-epresent a new species, the third known, of the remarkable genus
Rhinopithecus, and was described as follows : —

Rhinopithecus brelichi, sp. n. (Plate XXI.)
Size very large, apparently larger than either R. roxellance or
R. bieti ; for the skin, though that of a female, is as large as the
male of either of the other species. Fur not abnormally elongated
in any region, the longest being on the flanks, where the hairs
may attain to about 90 mm. in length ; those of the back 50-60 mm.,
and those across the shoulders 70-80. General colour of back
glossy slaty grey, the hairs grey to their roots, with shining tips.
A pi'ominent oval white patch, 5 inches long by 2 broad, present
in the middle line between the shoulders, its haiis white to their
roots. Crown sufiiised with yellowish, its hairs yellow at base,
whitening terminally, but with broad black tips ; haiis of cheeks

P.Z.S.1903,vol.I.Pl.XXI

STnit del.etlifh.

RHmOPITHECUS BRELICHI

Mon-ternBros . iirvp

1903.] MR. OLDFIELD THOMAS ox A ^'E^V ANTELOPE. 225

yellow with black tips ; nape between the yellowish crown and the
white wither-jjatch pale brownish with black tips to the hairs ;
neck both on sides and below blackish grey, the hairs dull whitish
basally, with black ends. Eai'S white, markedly conti-asting with
the head. Front of shoulders and inner aspect of foreai-ms deep
yellow, which shades into whitish along the under aspect of the
latter, and contrasts niai'kedly with the dark slaty of the outer
side of the foi'earin, this colour darkening to black on the wrists.
(The hands ai'e lost in the specimen, but are presumably black.)
Hind limbs light greyish, more or less suffused with yellow behind
and blackish in front, but the colour contrasts are not sharp and
defined as tliey are in the other species. Belly uniformly grey
(about grey No. 5 of Ridgway). Tail very long, conspicuously
longer than in the othei- species, its hairs, which average about
40 mm. in length, curiously curved on each side downwards and
away from the centime line, along which there is an ii-i-egular
parting ; in coloui' the tail is black thi'oughout except at the
exti-eme tip, whei'e there is a small white pencil ; on each side of
its base there is a small yellow patch, outside of which thei-e is a
blackish line passing round across the anal region : but owing to
the condition of the skin, the exact situation of these lines and
patches is not quite certain.

Appi'oximate dimensions of the type, measured on the skin,
which has been made up from a flat native pelt : — Head and
body 730 mm., tail 970 (with hairs 1040).

Habitat. Mr. Brelich states that, " as far as I could gather, this
monkey inhabits a. range of mountains known as the Yan Gin
Shan Range, about 108° E., 29"" N., in the north of the province
of Kwei-chow, Central China."

Type. Female. B. M. No. 3.3.14.1. Collected and presented
by Henry Brelich, Esq.

This magnificent Monkey, one of the lai-gest in the woi-ld apart
from the anthi'opoids, is a very i-emarkable discovei-y, and one on
which we may congratuLite Mi-. Bi-elich, who olitained and sent
it to the Museum on the suggestion of Mr. Herbert Ingram,
himself a frequent contributor to the National collections.

As may be seen from the above description and from the figure
(PI. XXI.), the differences between this monkey and its only near
allies are so numerous as to render any detailed com2:)arison un-
necessary. Good figui'es have been given of R. roxeUance by
Milne-Edwards ^ and De Winton ^, and of R. bieti by Milne-
Edwai-ds and Pousargues \

Mr. Oldfield Thomas also exhibited adult and young examples
of a Bush Duiker, which had been sent to tlie British Museum
by Mr. F. W. Isaac, from Eldoma Ravine, British East Afi"ica.

1 Rcch. Mainin., Text, p. 233, Atl. pis. xxxvi. & xxxvii. (1874). •

2 P. Z. S. 1899, p. 572, pi. xxxi.

3 N. Artb. MuR. (3) x. p. 121, pis. 9-12 (1898).

15*

226 MR. OLDFIELD THOMAS ON A NEW ANTELOPE. [Mar. 17,

It appeared to belong to a new species allied to the Congolese
Cephalophus iveynsi Thos., and was described as follows : —

Cephalophus ignifeu, sp. n.

Size medium. Fur close, fine and glossy, hairs of back about
an inch in length. General colour of back bright rufous or bay
(nearest to ochraceous-rufous of Ridgway), darkening forwards on
the neck and shoulders to dull brownish. Forehead mixed rufous
and black ; crown and occiput bright rufous like back, coronal
tuft a deeper and more chestnut or vinaceous rufous. Hairs of
occiput reversed upwards to the crest as usual, those of nape all
directed backwards. Muzzle blackish ; lips and chin white ; ears
dark bi-own behind, with white edges and inner surfaces. Throat
rufous. Belly brown mesially, grading into rufous laterally.
Inner side of forearms, inguinal region, and inner side of thighs
white. Outer side of forearms and thighs I'ufous ; feet brown,
darkening almost to black above the hoofs. Tail rufous above,
white below, proximally, with a mixed brown and white terminal
tuft. ,

Skull of normal proportions ; premaxillfe just reaching- nasals ;
frontal convex as usual ; posterior palate vaiiable, that of the
male cut out some way in front of the lateral notches, while in
the female the median notch is posterior to the latei'al ones.

Horns laid back just in the line of the face, those of an adult
male 98 mm. long, with a basal diameter of 31 mm. ; the same
dimensions in an immature female 49 and 20.

The young specimen (skull 105 mm. long, the last milk jDremolar
only just up) is nearly black all over, the coronal region and the
posterior back alone being rufous.

Dimensions of the type, an adult male, measured in skin : —

Head and body 810 mm., hind foot, with hoof, (c.) 240, ear 81.

Skull — basal length 162, greatest breadth 79*5, muzzle to orbit
94, nasals 73x32; palate length 97 ; length of upper tooth-row
53, of three upper premolars 23-4.

Hab. Eldoma R,avine, British East Africa, alt. 7200 feet.

T^/jue. Male. B.M. No. 2.11.17.6. Collected and presented by
F. W. Isaac, Esq. Three specimens.

Native name " Meindet."

This Duiker is allied to C. weynsi of the Eastern Congo and
C. johnstoni of Toro by its colour and general characters, but
difiers from them by the hairs of its nape being all directed back-
wards in the usual way. G. harveyi of Kilima-njaro, also a
member of this group, has a blackish forehead and no chestnut
coronal tuft.

The following papers were read :-

1903.] ON JAPANESE LONG-TAILED FOWLS. 227

1. Observations and Experiments on Japanese Lon<j;-taileJ
Fowls. By J. T. Cunningham, M.A., F.Z.S.

[Received January 1st, 1903.]

(Text-figures 41 ct 42.)

The enormous lengtli of the tail-t'eathei's in Japanese " Long-
tailed Fowls " has been known to zoologists in this country for many
years from the stutied specimens in the hall of the Xatui'al History
Museum in London. These specimens were figured by the late
G.J. Romanes in his 'Darwin and After Darwin,' Parti., puljlished
in 1892. The figui'es represent fairly accurately two males of the
breed, but the females are not represented at all, unless by indistinct
figures in the backgi'ound ; and the accessories, instead of being
di-awn from those actually present in the ^Museum case, are entirely
imaginary, showing the birds in a state of perfect freedom in open
country, a state which they never enjoy in Japan. Romanes's
figures are among a lai-ge number given as " typical proofs of the
eliicacy of artificial selection " ; and it is the object of the present
paper to show that this involves an assumption, with regard to
this particular breed, wdiich is by no means justified by the facts.

The tails of the male specimens in the National Museum are
6 to 9 feet in length, that is to say, the longest feathei's are of
that length. I have endeavoured, but without success, to obtain
information concerning the histoiy of these specimens. I l)elieve
they were reared in Japan, and they were probably sent fi-om that
country after death, either as skins or as stufied specimens.
There are some female specimens also, but in these the tails are
scarcely longer than in the liens of ordinary bi'eeds.

The breed has been known to poultry-fanciers for a considerable
time, and the following is the account of it given in ' The Book of
Poultry' by Lewis \Yright, published in 1885 : —

" About the year 1878 there appeared in CTermany, and a year or
two later in England, fowls impoi-ted from Japan, wdiose principal
peculiai'ity consisted in an immense length of tail and hackle-
feathei'S. Some of these were exhibited as Yokohamas ; others,
said to be superior iii these points, were called Phoenix fowls.
The tails of these specimens averaged about a yard in length, and
the genei'al appeai-jince was not only that of a Game-fowd, but all
the colours were Game coloui'S, Whites, Piles, Duckwings, and
later a few Black- Reds. The long plumage Avas, however,
unique, and a fair idea of it may be gathei'ed from the illus-
tration.

"Correspondence in the poultry journals brought out the fact
that such birds had been occasionally exhibited as Japanese game
.so far back as about 1872. But it further appeared that in the
Japanese Great National Museum at Tokio there were pi-eserved
two specimens, of an allied race in which the tail-feathei'S measure

228 MR. J. T. CUNNINGHAM ON [Mar. 17,

13| feet and 17 feet respectively, and a feather lias been actually
sent to France which measures 2 metres 85 centims. in length
(say 9| feet). In 1884 Mr. Gerald Waller, of Twywell, imported a
pen of these biixls ; and from his statements we gather that they
are known in Japan as Shinowaratao, Shirifuzi, or Sakawatao
fowls and various other names. He says the very long-tailed
ones are kept in high, narrow cages, always sitting on a perch
covered with straw- rope, with no room to turn or get down, hut
with a food- and water-tin at each end of the perch. Three times
daily they are lifted down for a few minutes' exercise, their tails
being carefully rolled up in paper cases to keep them from injury.
The Japanese state that a tail has been measured 23 feet in length,
and that the birds on\j moult the tail once in three years. This
last statement is highly interesting. It is obvious that if a tail
23 feet long were grown in one yeai-, it must be at the rate
of nearly three-quarters of an inch per day ; and though
Madame Bodinus states that she could see the tails grow daily,
it is difficult to realise this ; but experience will soon decide the
point. The birds which have reached Europe have never yet
exceeded 5 or 6 feet in length of feather, which is not beyond the
possibility of a single season, though it appears of an enormous
length. The saddle-hackles of Mr. Waller's birds are 16 inches
in length ; but it is manifest that such enormous feathers as
reported from Japan covild never be preserved under the ordinary
conditions of an English poultry-yard. The feathers are not only
long but extremely nai'row and flexible, trailing low after the
birds."

Mr. Wright does not mention the comb, but the illustration
which he gives represents the male bird with what is called a pea-
comb of small size and with small wattles, whereas the specimens in
the National Museum have single vei'tical serrated combs and large
wattles. The pea-comb is a rounded mass, with small rounded
tubercles projecting from it.

Mr. Frank Rice, of Acton, Sufiblk, who breeds Yokohamas,
gives in his circular the same illustration which appears in
Mr. Wright's book, and which therefore is certainly not a new
figure from a living specimen. But in his description he states
that the head should be neat and small, with evenly-set pea-comb.
It would thus appeal' that the long-tailed fowls coiiiiDrise varieties
which differ in comb as well as in colour, though they seem to be
all similar in the excessive growth of the tail, and probably are
all grown in Japan under the same artificial treatment.

The principal purpose of this paper is to discuss the causes by
which the elongation or excessive gi'owth of the tail has been
produced. About two years ago, Pi'of . Lankester, in a letter to
' Nature,' refeired to the specimens in the Museum of which he is
Director as " a magnificent sport," comparing their exceptional
character to what is called genius in human beings. On the
other hand, in the ' Dictionary of Birds ' by Newton and Gadow,
article " Feather," the length of the tail-feathers is attributed to

1903.1 JAPANESE LOXG-TAILED FOWLS. 229

continuous growth, and it is stated that the moult is checked or
prevented by some means unknown to Europeans. We have seen
that in Wright's book the statement is cited that the feathers are
only moulted once in three years, but this might be a congenital
peculiarity, whereas the expression used in the ' Dictionary of
Birds ' implies that the moult is prevented by artificial treatment.

Yery definite statements on the question are made in a paper
by INIr. Basil Hall Chamberlain (" Note on a Long-tailed Breed
of Fowds in Tosa," Tiuns. Asiatic Soc. Japan, vol. xxvii. 1900),
for my knowledge of which I have to thank^ Mr. Frank Finn.
Mr. Chamberlain made enquiries on the subject in the country
from which the breed is obtained, and where presumal)ly it
originated, but nevertheless his statements are based on the
assertions of l)reeilers and not on any investigations of his own.
The following is a summary of his paper : —

The origin of the Ijreed is not knowai, Init it is believed to be at
least a hundred years old. It has been produced simply by
selection of the best specimens ; one highly-prized variety, the
Haku, was produced in this way within the last few^ yeai-s.

The proper general name for the Long-tailed fowds is Shino-
wara-to, derived from the village of Shinowara in the province of
Tosa, east of Kochi the capital. Some are still bred in that place,
but most now in Kochi itself, whence the majority are exported
to Kobe and some of the finest to Tokio, but the very finest are
I'etained by the producers.

The following varieties were described to Mr. Chamberlani : —
Shira-fuji : white head- and body-feathers, tail black as in the
other vai-ieties. He saw one specimen of this two years old, and
measured its tail-feathers, which were 7^ feet long. Another
specimen, fourteen months old, had tail-feathers 4 feet long.
Others have black bodies. Another variety is the Haku, white
all over with yello\v legs ; another, Totenko, has red neck- and
body-feathers ; another, Dokiri, has reddish colour mixed with the
white of the body. All these except the Haku have black tail-
feathers. , 1 . 1

As great a length as 18 feet has been reached m the tail-
feathers, but even 12 feet is a rarity. From 7 or 8 to 11 feet is
the usual length. The feathers grow about 4 inches a month,
and continue "to grow while the Inrd lives, which may be eight or
nine years. The beautiful body-feathers growdng from the
shoulders reach a length of 4 feet. (iTo^e.— This evidently refers
to the saddle-hackleSjVhich grow^ not from the shoulders but from
the top of the rump.)

Some of these saddle-hackles may fall ofl" in moulting, but the
tail-feathers never do so. He saw the birds in Octol)er 1898
when moulting, and only tlie ordinar}' feathers were gone or
going, not the long ones.

He also saw the hen, which was a handsome Inrd, distantly
reminding one of a hen-pheasant, with fawn-colouretl breast and
wdiite quill to the delicately-coloured feathers of the back. She,

230 MR. J. T. CUNNINGHAM ON [Mar. 17,

too, has longer tail-feathers than an ordinary hen, sometimes as
long as 8 inches. One, or at most, two hens are allowed to each
breeding- cock. The latter's tail-feathers are cut to allow of his
walking about freely. He lives a little longer than the others
which must be kept shut up ; but all are hardy, bearing both heat
and cold. The ordinary number of long tail-feathers is 15 or 16,
some cocks have as many as 24.

The tail-feathers must not be wound up, as people ignorantly do
away from Kochi, but must always be allowed to hang free ; for
which reason the cocks are kept in high, nari'ow cages, quite dark
except at the top, for light at the bottom would attract them.
When the tail-feathers become too long and touch the ground in
the cage, a bamboo is put a little way back so as to form an arch
and make more distance. The birds sit all day on a flat perch
3 inches wide, and are only taken out once in two or thi'ee days
and allowed to walk about for half an hour or so, a man holding
the tail all the while to prevent its getting torn or soiled.

The high, narroAv cages may be made of any wood ; they are
6^ feet high, 3 feet deep, and 6 inches wide. The wonderful
feathers both on tail and body come from quills much stouter than
any possessed by ordinary fowls.

The price in Kochi was 15 dollars for a cock with tail under
10 feet, 25 dollars over that length.

There is absolutely no artificial method of making the feathers
grow. All is done by selection. Any failure is due to not having
a hen or parents of the proper breed. Also one must know how
to treat the birds.

At Kobe in November 1898 Mr, Chamberlain saw three speci-
mens, one with tail-feathers 13| feet long. He also saw a splendid
Avhite tail 10| feet long, which had been pulled out from a white
bird owing to its falling off its perch and fluttering about. The bird
was five years old, and the feathers were growing again. The fancier
said that the feathers in young birds grow about 4 inches a month,
in older birds more, up to 7 inches a month.

Two photographs are given with the paper, but no reference to
them is made in the text. The tails are very long, but there are
no long feathers from the shoulders, only tail- and saddle- or
rump- hackles.

It is evident that Mr. Chamberlin, although his observations are
of considerable value, was not an experienced naturalist, and that
he is simply reporting what he was told. He says nothing about
the combs of the fowls he saw. With regard to his assertions
about selection as the sole means by which the breed has been
produced, it is to be noted that he is evidently referring in the
case of the Haku chiefly or entirely to colour. The Haku is a
white variety, and Mr. Rice, above mentioned, also has a strain
of this colour. Nearly all domesticated birds and animals vary in
colour, and nothing is easier than to separate a white variety in
fowls, horses, pigeons, dogs, &c. These fowls, like others, vary
in colour ; and the question before us now is not the separation of

1903.] JAPANESE LOXG-TAILED FOWLS. 231

colour varieties, but the excessive length of the tail which occurs
in all the vaiieties. Mr. Chamberlain states that a specimen
fourteen months old had a tail 4 feet long. At the rate of
4 inches a month, this would mean a period of growth of twelve
months, and therefore the chickens must have begun to acquire
their permanent feathers at two months of age, and the fejxthers
must have grown without intermission afterwai'ds. This is
pei-haps not impossible, but accoi'ding to my observations, recorded
below, it is very improb;iiile. It may further be pointed out that
the statement " there is absolutely no artificial method of making
the feathers grow," is ditiicult to reconcile w- ith that which follo\vs
it, " also one must know how to treat the binls,"' unless proof is
offered that the artificial treatment has no efl'ect on the growth of
the feathei's.

The birds in my possession, which have formed the subject of
my own observations and experiments, were descended fi-om a
pair which were imported direct from Japan by ^\v. John Sparks,
of London, and purchased by ]\Irs. J. C. Williams, of Caerhayes
Castle, Cornwall. Before proceeding to my own observations, I
wish to discuss the question Avhether the moult occui'S in the
specimens kept in this country, and also some evidence I have
obtained as to specific treatment of the feathei'S. We have seen
in j\Ii-. Wi-ight's account that the feathers of bii'ds in this countiy
had not been known to exceed 5 or 6 feet in length. This
does not thi'ow much light on the question of the moulting,
because it may be that, the bii'ds not being kept on perches as in
Japan, the feathei'S get bi'oken when they have reached a, con-
siderable length. I enquii'ed of j\Ii'. Rice what his expeiience
was, and he replied that his bii-ds usually moulted their tail-
feathei'S each year, but he had had some cocks omit this operation
in their second yeai-. He said that his cockerels at eight months
of age had tails from 2 ft. 6 in. to 3 ft. 6 in. in length, and that
he had one cock, three years old, that had a tail 5| feet long.

Mrs. J. C. Williams was also kind enough to answer my
enquiries. She informed me that the original male bird which
she obtained from Mr. Sparks had a tail just 5 feet long. At first
its plumage suffered from the change of climate, l)ut in 1901 its
tail Avas considerably better than it had been. (The length at
this time was not stated.) She had a young bird whose tail had
measured 4| feet, but he broke it in getting about. The young
birds usually moulted about a yeai' aftei' hatching. This must have
been the first moult of the adult plumage, which begins to appear
in the autumn aftei' hatching. In reply to a request for fuither
particulars about the moult, ]\rrs. Williams stated the birds do
certainly cast their long tail-feathers. The bird that had some
feathers 44 feet long was then (INIarch 1902) nearly three years
old, and she thought it Avas at the second moult that a feather of
that length was measured.

There is evidence here that, when left to themselves, the long
tail-feathers are moulted in the ordinary way, at any rate for two

232 MR. J. T. CUNNINGHAM ON [Mar. 17,

successive seasons after the first year in which the birds are
hatched.

It will be seen that the absence of the moult, or the occurrence
of continuous growth, if true for the birds in Japan, is not true
for those reared in this countiy ; and thus we have i-eason to
doubt that uninteiTupted or continuous growth of the tail-feathers
is a fixed congenital peculiarity in the breed.

Mr. John Sparks in May 1901 supplied me with the following-
information concerning the method of treatment applied to the
birds in Japan : — ^" In ordei- to ensure veiy great length of tail,
the cocks ought to be kept on a perch as much as possible after
they are six months old ; and the tail-feathers should be pulled
gently every morning, grasping the centre bone-like part firmly
with the fingei' and thumb and pressing steadily downwards
towards the tip, each feather being done several times. This
softens the quill and causes it to lengthen. The birds do not
moult the tail-feathers, but if one or more come out others
immediately grow in their places.

" The Japs themselves, those who take great pride in their birds,
always roll the long feathers up like a lady rolls up her hair, and
tie them, whenever the birds are let off their perches to walk
about, which is about twice a day for an hour at a time.

" I have often seen them thus treated in Japan, and those
which Mrs. Williams and the Hon. W. Rothschild had from me
were so treated on the voyage by the man in charge of them, and
I sent them down to St. Austell in their regular perch-cages."

Thei'e is here a detail in the treatment of the feathers which,
so far as I can discover, has never been mentioned in any
published account of the matter ; and my own experiments, which
I now proceed to describe, so far as they have yet gone, tend to
show that this mechanical treatment of the feathers is the whole
secret of the mystery.

My own Experiments.

In May 1901, I received from Mr. John Sparks twelve eggs,
laid by the fowls of this breed in the possession of Mrs. J. G.
Williams. I afterwards ascertained that some of these were the
produce of the pair originally imjDorted from Japan, others were
from the offspring of this pair, but all the eggs were of perfectly
pure breed with no cross or admixture whatever. The eggs were
incubated by an English hen of mixed breed, and on June 13th
ten healthy chicks were hatched. Of the other two eggs, one
contained a dead chick, the other was addled or not fertile. This
shows that the eggs were remarkaljly fertile and of great vitality.

The chicks were small, the eggs being not much more than half
the size of ordinaxy table-eggs. Their colour was fawn, with a
broad dark brown stripe down the middle of the back, a narrower
stripe on each side of this, and a thin stripe of the same colour
rimning from the outer corner of the eye. They were very

1903.] JAPANESE LOKG-TAILED FOWLS. 233

pretty, very active, and very lieixlth}-, and tlii'ove well on ordinary
food, coiLsi.sting of oatmeal, chopped meat and vegetables, and
mixed grain of vai-ious kinds. One of the chicks was accidentally
killed when I turned the hen out of the nest, having got beneath
its mothei-'s feet as she was scratching the earth. I examined
this specimen, and found that the primaries and secondaries of
the wings were pi-esent as short black quills with a little down at
the tip, l)ut that there was no trace of tail-feathers or tail-coverts,
nothing but down over the rest of the body. The comlj was
visible at the l)ack of the beak as a slight yellow ridge with six
teeth. The toes were four in number ; the skin on the legs
was yellow.

Jane 23;yZ. Age 10 days. — Another chick had been lost u[) to
this date, having escaped and probably fallen a ^•ictim to a cat.
The feathers of the wings now i-eached nearly to the end of the
body, and wei-e chequered in colour, l^eing mai-ked ti'ansversely
with dark colour and gi-ey alternately. The tail-feathers, i. e.
rectrices, had begun to sprout in some of the chicks, elsewhere
thei'e was still only down. The combs ajipeared no larger.

Jahj Gth. Age 23 dai/s. — The chicks were now half-fledged :
anothei- had been lost by escaping, so that only seven survived.
Feathei's wei'e growing on the shouUlers and sides of the bi'east,
but the head, back, and fi'ont of the breast were still downy, with
the original markings. Tail-coverts also appearing. Feathers,
except tails, all barred with dai'k and grey, so that the general
appearance was speckled and veiy inconspicuous against the soil.
The tail-feathei'S were dark, near!}' black, Ijut not steel blue-black
like those of the adult. It was interesting to see the chicks scatter
in all tlii-ections, and then ci'ouch down whenever the hen uttered
her special warning ciy, as she did often when she heard a jackilaw
ci-oak. In four of the chicks the tails were more developed than
in the rest ; in these also the combs were beginning to grow
higher and to get red. These foui' pi-oved to be cocks, so that the
sexual difference begins to show itself at this eai-ly age.

Jiobj 28th. Age 1 month 15 dai/s. — ^The four cocks showed
their sexual charactei's a little more distinctly, the wattles and
ear-lobes being indicated by a tinge of I'ed. In the three hens
the combs had not l)egun to enlai-ge and showed no i-ed.

One of the cocks was darker than all the other chicks, and had
a slender tail, not very long and l)ent downwai-ds. The hens were
light -coloured, with white bi'easts, their tails being as long as those
of some of the cocks. Two of the cocks had dai-k breasts.

Aiog. 4ith. Age 1 vwnth 3 toeeks. — Chicks now fully fledged.
In the four cocks reddish tints were appearing in the feathers of
the back and wings, while in the hens only neutral tints were
present. The combs in the cocks were a little more developed, in
the hens not developed. The tnil- coverts in the cocks were growing,
but not longer than the rectrices, and the sickle-feathers not
conspicuous.

The red feathers mentioned' above are the beginning of the

234 ME. J. T. CUNNINGHAM ON [Mar. 17,

adult plumage. There is at least one complete change of feathers,
from the chicken plumage to the adult, both in cocks and hens.
Whether the chicken-feathers are changed before this I do not
know, but there were indications that the wing- feathers at least,
in the chicks, had been changed before the assumption of the adult
plumage.

Aug. 12th. Age 2 months. — I examined one of the cocks, and
could distinguish two steel-blue sickle-feathers on each side, not
longer than the rectrices and rather narrow. These were the first
of the adult tail-feathers to appear, succeeding others of dark,
lusti'eless, bi'ownish black.

Aug. 20th. Age 2 months 1 iceek. — I counted the tail-feathers
in some of the chicks. In one of the hens I found 14, or 7 pairs,
of rectrices. In one of the cocks, the dark one previously men-
tioned, the saddle-hackles were almost black with a grey stripe
down the centre, instead of reddish. In this bird I found also
seven pairs of rectrices, and only four pairs of sickle-feathers
which were neuti-al bi'own. The other three cocks could also now
be individually distinguished, and I give the peculiarities to show
the amount of variation : —

A. The largest : more advanced in plumage than the rest, more

red on the back, breast nearly all black except at the sides ;
a kink in the comb.

B. Similar to A, but not quite so far advanced.

C. Similar to A and B, but lighter in tone ; back rather

yellowish than red.

D. The dark bird above mentioned.

Au,g. 2Ath. Age 2 months 11 days. — Examined Cock A and
found a number of steel-blue sickle-feathers, or tail-coverts, all
with very long horny sheaths at the base indicating vigorous
growth. In Cock B similar feathers were only just beginning to
sprout.

Aug. 2&th. — In Cocks C and D very few new sickle-feathers
showing. In B the rectrices were being replaced.

Sept. Qth. Age 2 months 3 loeehs. — In Cock A all the chicken-
rectrices had been moulted, and the new steel-blue ones were
growing out with long horny sheaths at the bases. In all the
cocks the sptirs had begun to show as very slight blunt knobs.
ISTone of the cocks had crowed yet.

Sept. 10th. — Examined one of the hens. Saw several tail-
coverts growing with long sheaths but no new rectrices.

In Cock D, the dark one, the two central rectrices were long,
slender, and curved, and had sheaths at the base, also sheaths to
some of the other rectiices. I thought at the time that these
central rectrices had not replaced others in the chicken plumage,
but grown continuously. I afterwards concluded that they had
moulted earlier than in the other cocks.

In Cock A the longest tail- covert measured 19 cm. or about
7 1 inches.

1903.] JAPANESE LOXG-TAILED FOWLS. 235

tie-pt. 1 9i/i. 2i(je 3 months 1 loeek. — Condition of plumage : —

Cock A. Breast, thighs, and l)elly all black. Xeck-hackles
very light grey, with thin dark stripe down the centi-e of
each feathei'. Back with some steel-blue feathers behind
the hackles, the rest red. Saddle-hackles developing,
yellow, long and thin. Nearly all the rectrices and tail-
covei'ts growing with long sheaths, colour steel-blue.

Cock B. Nearly the same as A, but not quite so far advanced.

Cock C. Much less advanced, white on sides of breast, speckled
brown feathers mixed with red of back, only the outer-
most rectrices shomng sheaths.

Cock D. Very little white on breast, hackles of neck steel-blue,
back black with a little red at tips of feathers. Saddle
whitish. Rectrices 7 pairs, only the central and outer-
mo.st paii-s with sheaths, the rest appai-ently not yet
moulted.

Hens. Breasts almost white, the sides of In-east bufi"; neck-
hackles dai-k, with whitish stiipes down centres of f eathei'S.
Back and tail greyish bi-own, speckled, i. e. with white
quills. Two hens with black heads, one Avith head lighter,
grey and speckled.

I now decided to stroke and pull the tail-feathers in Cock B and
to leave Cock A untouched, to see if any difference would result.
These two, as I have said, were closely similar except that A was the
larger', finei- biixl and slightly more advanced in development. I
chose, thei-efore, the one which w'as congenitally inferioi-, so that
if an}'- superiority in growth of feather appeai-ed in it, it could
only he due to the artificial treatment. In both thei-e wei'e seven
pairs of rectrices, the central pair slender and curved, the rest
broad and straight. The outermost paii- had not yet been
moulted. I fixed up a sort of cage with a round pei-ch at the
bottom, and put Cock B into it while I stroked his feathers, but
did not keep him in it.

Oct. \st. — Found that the hens had shed most of their rectrices
and were producing new feathers, as well as new tail-coverts, liut
without change of coloiu-.

Oct. 1th. — The ui^per tail-covert in Cock A measured 25 cm. or
about 10 inches, not including the basal sheath.

At this time I changed m}' r-esidence, and the fowls were
installed in a place divided into two runs ; into one I put Cocks B,
C, aird D, in the other Cock A with the three hens.

On Oct. 12th Cock A crowed for the first time, another proof
that he was a little more precocious than the others.

Oct. \^th. Age 4 months 3 c?«..ys.— Longest feather, a tail-
covert on left side, in Cock B 27 cm., about the same as the
longest in Cock A.

Oct. '27th. Age 4 months 2 loeeks. — Tried tying the cocks by
one leg on ordinaiy perches about 3 feet from the ground, and
thus was able to measure the feathers better.

236 MR, J. T. CUNNINGHAM ON [Mar. 17,

Cock A. Longest feather 13 inches from skin.

Cock B. The same. But there were 7nore long feathers in
Cock A, the outer ones being more nearly the length of the
middle.

As I found I had not time to pay attention to more than two
cocks, and as, moreover, they soon began to fight, I got rid of
Cocks C and D ; and what I have to say hei'eafter refers only to
Cocks A and B, which were undoubtedly the best of the four, and
had the most beautiful coloui's. I began tying these two cocks
on ordinary perches in the daytime by means of a piece of tape
tied to one leg and round the perch.

Bee. 1st. Age 5 tnonths 18 days. — Longest feather in Cock A
18 inches, in Cock B the same. This shows a growth of 5 inches
in five weeks, oi- 1 inch per week. Pei'haps the feather in Cock A
was really a little longer, as neai'ly all the feathers in this bird
had lost their tips by friction against the ground. There is thus
no evidence up to this time that pulling the feathers in Cock B
has increased the rate of growth.

Dec. 8ih. — I had been stroking the feathers of Cock B regularly
at morning and evening. Usually I tied the birds on the perches
at night, and generally tied the feathers up in paper when the
birds were free. I used tissue-paper, and rolled the tail up trans-
versely, fastening it with tape. I tried a rough naiTOw cage with
wooden bars foi' Cock B, but the bird turned round and escaped
through the opening behind the perch.

I noticed now that one of the long feathers in Cock B seemed
to have stopped growing, the sheath having peeled ofi" and left a
dry stalk. This was one of the outer tail-coverts, but the rest
were still growing vigorously.

Dec. 27th. — As my first attempt at a cage did not succeed, and
T was unwilling to confine the birds so completely as the Japanese
do, I made a. perch of about 9 inches in length supported on two
uprights. This apparatus was movable, not fixed to the ground.
I used this for Cock B, tying him to it by a piece of tape fastened
round one leg. This answered veiy well, although he once upset
the perch and slightly wounded his head. No permanent injury
resulted from this accident.

On this date, when I was sti'oking the feathers, one of the
smaller tail-coverts on the right side came out of the socket, with
the epidermic sheath attached to it. This feather was 14 inches
long. This seemed to indicate that the eflfect of stroking the
feathei's was, as suggested by a correspondent of ' Nature,' to pull
the gi-owing feathers out, instead of increasing their growth ; but
my experiments were not finished yet.

Dec. Slst. Age 6 months 18 days. — Measured the feathers as
accui'ately as possible on a flat wooden measure.

Cock A longest two feathers, the central feathers of the tail,
22 inches.

Cock B longest feather scarcely 22 inches, also a central
rectrix.

1903.] JAPANESE LOXG-TAILED FOWLS. 237

It seemed, therefore, that the gi'owth of the feathers in Cock B
was barely keeping pixce with that in Cock A, although only in
the former were the feathers pulled or stroked. In Cock A the
longest feathei's Avere the two central i-ecti'ices ; in B only one of
the central rectrices was as long, the adult feathers having mostly
staited earlier in A than in B.

Up to this time Cock A had been with the three hens, and B
with the other cocks until I had got rid of these.

1902, Jan. 8th. — On this date one of the hens laid the first egg.
It was small and of light brown colour, like those the birds were
hatched from.

The hens have a low seii'ated comb and scarcely any wattles,
only red skin about the cheeks.

Ja7i. I2th. Age 7 months. — In Cock A the 2nd pair of rectrices,
counting from the middle, seemed to be coming to the end of their
growth, the sheaths diying up and the quills forming. This pair
wei-e not so long as the corresponding paii" in Cock B which
had been pulled. The five outer pairs of rectrices had ceased
growing in both cocks. These were bi-oad and stiff and only about
6 or 7 inches long, in fact like the i-ectiices of ordinaiy fowls.
All the hens were now laying. One of the hens was put with
Cock B, the other two Avith A.

Feb. Ath. Age 7 months 22 dai/s. —

Cock A, longest feather 25i in. = 64 cm.

Cock B „ „ 25| in. = 65-7cm.

It is thus evident that the gi-OA\i;h in B had been greater than
in A. The feathei'S in B wei-e pulled once or twice a day ^\•hen-
ever I was at home, and B was tied on the pei-ch with his feathers
loose at night, while A was usually left free with his tail I'olled
up in papei-. The sheaths of the centi-al i-ectiices, which were
the longest feathei-s in both birds, wei'e in Cock A shoi-ter, and
seemed as though gi'owth were about to cease.

It is to be noted hei-e that so long as the feather is in full
growth, the horny sheath which surrounds the base is milky
white, or bluish white, and soft, while when growth ceases the
sheath dries up and appears black because the feather is seen
through it ; then it scales off, and leaves the quill of the feather
bai'e down to the surface of the skin. Tlie quill at the base of a
full-growai feather is only foi-med at the end of growth, and most
of it is contained in the skin-socket. Thus I leai-ned to I'ecognise
the approaching cessation of growth and formation of the (piill by
the appearance of the horny sheath.

Feb. 10^/i.— One of the feathers was accidentally jiulled out in
Cock A while the tail was tied up in paper. I do not know how
this occurred ; I only saw the I'oot of the feather hanging free
while the length of it was held by the paper. Wlien a gi-owing
feather is thus pulled out, the horny sheath comes out of the
deimal socket and i-emains attached to the base of the feather,

238 MR. J. T. CUNNINGHAM ON [Mar. 17,

and the end of the sheath is quite soft and exudes a little blood
and moisture when pressed. This feather was 52-8 cm. or about
1 ft. 9 in. long. It was one of the anterior tail-coverts, and was
the only feather ever pulled out of Cock A.

Feb. Wth. Age 8 months nearly. — On this date, when I was
stroking the feathers of Cock B, another came out. It w^as one
of the anterior lateral coverts of the right side, and its length was
15^ inches (38 cm.). The feather had not ceased growing, but
showed signs that it was about to cease, as the rhachis at the base
was stout and thick, and the barbs emerging from the sheath were
free and downy, as they are at the base of a full-grown feather,
and never at its more distal part

Feb. \ith. — Examined the bases of the feathers in both cocks.
In A the central rectrices had nearly or quite ceased growing, and
also nearly all the tail-coverts, only two or three still showing the
basal sheaths. In B two or three of the posterior coverts had
ceased growing on each side, but a number of the others still
showed vigorous growth.

Feb. I5th. — The longest feather in Cock B came out when I
was stroking the feathers, though I was not pulling at all hard.
Tlie feather was the central rectrix of the left side, and was in
growing condition, with bluish pulpy sheath at the base. It came
clear out of the socket and was not broken off at all. The extreme
length was 2 ft. 41 in., but about 1 inch of this was in the socket
before it was pulled out. The total length of the sheath up to
the point where the feather proper emerged was 3-1- inches (8 cm.).

Feb. IQth. — Ascertained that in Cock B three tail-coverts on the
right side had ceased growing and two on the left. The 2nd
recti'ix on right side had ceased to grow, that on the left nearly so.

In Cock A the 2nd pair of rectrices had ceased growing, the
1st or central pair had nearly ceased. Two central coverts, which
were nearly as long as the central rectrices, had nearly ceased, and
there were only two others on each side with growth-sheaths.

There were two rows of coverts in front of the rectrices, anterior
and posterior, with four or five feathei\s in each i-ow on each side.
When the feathers are specified individually they are counted from
the middle line.

Feb. nth. — One of the posterior lateral coverts on i-ight side in
Cock B was pulled out today. I was really trying to see if it
would come out when pulled, as it had completely ceased to grow.
I found that it required much more force to pull it out than is
required in the case of a growing feather. The quill was com-
pletely formed. It was evident that the attachment of the feathers
was most feeble when growth was ceasing, not when the feather was
growing vigorously ; while after the quill was completely formed,
the feather was very firmly held in the socket. The feather was
19| inches long.

March 2nd. — In Cock B another feather was unintentionally
pulled out. It was in full gro^vth, and was one of the lateral
anterior coverts on the right side. Total length 16^ inches.

1903.] JAPANESE LONG-TAILED FOWLS. 239

March 4i/i,— Mejisui-ed feathers, exactly four weeks since last
measurement : —

Cock A, longest feather 2 ft. 3.J in.

Cock B „ „ 2 „ U „

The longest feathers in A were the two central rectrices, and
these had almost entirely ceased growing since last measurement.
They had grown 2| inches in four weeks.

In B, on the other hand, the longest feather was the central
rectrix on the right side, wdiich was still growing and was 1 inch
longer than the rectrices of A. The left rectrix of B, which was
pulled out, would have been somewhat longer.

In Cock A there were now only two feathers with growing-
sheaths, both on the right side, all the rest having ceased to grow.
in Cock B, on the other hand, there were several coverts with
growing sheaths, as well as the central rectrix of the right side.

March 9th. — The two feathers still growing in Cock A were the
central posterior covert on the right side anil one next to this on
the same side. They seemed to be about to stop growing. There
wei-e only five long feathers in Cock A, all the rest being shortei-
than the coi-i-esponding feathers in B.

March 10th. — Another featlier came out fi^omCock B today when
the feathers were stroked ; it was the 1st anterior covert on the left
side, and, although still growing, showed signs of cessation of growth.
It was evident that the featliei-s, as noted abo\-e, were easily
pulled out at this stage. Feather measured 1 ft. 9| in. (55-3 cm.).

March llth. — Still another feather pulled out of Cock B, 1st
anterior covert of right side. It measured 1 ft. 8 in.

March 16th. Age 9 months. — The two feathers which were
still growing in Cock A seemed now to have both ceased to grow,
one of them certainly. The only other growing feather in this
bird was the one on the left side which had sprouted in place of
the one pulled out.

In Cock B four of the old feathers weie still growing, namely,
the right centi-al rectrix, and thi-ee coverts on the left side, besides
the new feathers sproiiting in place of those pulled out.

Many of the golden saddle-hackles in B had growing sheaths.
One came out when they were being stroked the other day, and
measured 7 inches in length. A few of the feathers of the same
kind were growing in A, but most have stopped, and these hackles
altogether were about 2 inches shorter in A than in B.

Between the saddle-hackles and the long tail-coverts there were
a number of featheis which may be called transition feathers.
They were flexible and projected upwards first and then drooped
in an arch ; the outei- ends were steel-blue, the bases cariied pure
white down, which was exposed and was very ornamental to the
bird.

Aj^'il 1st. — Measured the feathers, exactly four weeks since last
measurement : —

Longest feathers in A, central rectrices, 2 ft. 4j in.
Longest feather in B, right central rectrix, 2 ft. 8 in.

Proc. Zool. Soc— 1903, Vol. I. No. XVI. 16

240

MR. J. T. CUNNINGHAM ON

[Mar. 17,

The rectrices in A had therefoi'e grown 1 inch in four weeks,
the rectrix in B 3| inches. The former had ceased to grow some
time before this date.

One coveit on the light side in A had still a httle of the blue
sheath at the base ; it had almost finished growing, and was still
shorter than tlie four longest central feathers.

In B the next longest feather was the central covert of the left

Text-fig. 41 .

Japanese Loug-tailed Fowl.
Coct A, photographed April 1903.

side, which was 1 ft. 1 1 in. long, or 9 inches shorter than the central
rectrix. The left central rectiix, which had been pulled out on
Feb. 15th, was growing again rapidly and was now about 4 inches
long.

In Cock B the left central covert Avas still growing vigorously,
with long sheath. It had a twist in it and hung in a spiral. Of
the other two growing coverts on the left side, the outer had
nearly ceased growing, the other still had a blue sheath.

1903.

JAPANESE LOXG-TAILED FOWLS.

241

1 found that tlie featliers wliicli lind ceased <i;rowiiiy for some
time were very firmly attached, it was impossible to pull them
out by moilei'ate force, and I did not wish to pull them out by
violence. Tliei-e were 8 lonjj^ coverts on each side in both cocks.

April 7th. —There were siii^ns that more of the feathei's in Cock B
were coming to an end of their growth. The sheath of the right
central rectrix was diminishing. The outer of the two n[)per

Text-fig. 42.

.lapanese Long-tailed Fowl.
C;o(,k B, photographed April 1908,

posterior coverts on the left side had, 1 believe, ceased to grow,
and the inner, though still growing, was appai-ently aliout to cease.
The only feather which appeared to be in full growth was the
central covei-t on the left side.

April VMh. Aye 10 months. — The right central rectrix in
Cock B came out today wdien I was pulling the feathers. I had
pulled it I'ather hard and rathei' frequentlv in the hope of stimu-

16*

242 MR. J. T. CUNNINGHAM ON [Mar, 17,

lating its growth, with this result. I am sure it would not have
grown much more, as its growth was evidently ceasing.

Total length of the feather 2 ft. 9§ in., of which | in. was
embedded in the socket.

April 20th. — On this date the light- coloured hen being bi'oody,
I put 9 of her own eggs under her to be incubated.

April 29fA. — Longest feather in Cock B now was the central
covert of the left side, which was 2 ft. 2 in. long and still gTOwing.
The covert next to this was also growing. The feathers which
had been pulled out were all regrowing vigorously, the longest
being the left central rectrix, which was 7 or 8 inches long.

In Cock A there was no further growth, except in the anterior
covert which had been pulled out.

May 1 1th. — All of the eggs incubated by the light- coloured hen
were hatched on this date, and all the chicks were vigoi-ovis and
healthy. My experience of the breed has been in direct opposition
to the statements in Lewis Wright's account, with regard to their
vigour and fertility : I have found them exceptionally hardy,
vigorous, and fertile.

There was no new variation observable in the chicks, except in
one which was a rich golden-brown on the head and shoulders,
instead of light buff colour. All were striped as described in the
original chicks of the preceding season.

May 27th. Age 11 months 14 days. — Left central covert in
Cock B 2 ft. 5 in. long, so that it had only grown 3 inches in four
weeks. The covert next to this came out on this date when I
stroked it. It was 2 ft. 2| in. long. It had evidently nearly
ceased to grow, and was beginning to form downy barbs at the
base.

June %id. — First signs of moulting noted in the hens ; a good
many of the breast- and body-feathers on the grovmd from day to
day, and one primary wing-feather from the light- coloured hen
found. There was, however, no sign of moulting in the tail-
feathers of either cocks or hens. Cock A had lately been left
at liberty with tail free, and none of the feathei"S had been
broken or lost.

June Sth. — Hens moulting a good deal, moi-e primaries shed,
and in some cases the new primary was half developed. The
primaries in each hen were shed a single pair at a time.

In the cocks none of the wing-feathers, saddle-hackles, or tail-
feathers had been shed.

The chick of this season's brood which was browner than the
others in the down, showed the same peculiarity in the immature
plumage, having a yellowish -brown colour instead of the neutral
brown of the others ; the marking was the same.

Ju7ie 24:th. Age 1 year 1 1 days. — Left central covert in Cock B
2 ft. 8 in. long, so that it had grown only 3 inches in four
weeks.

July I3th. — The left central covert in Cock B was accidentally
pulled oiit when I was away fiom home ; the paper in which the

1903.] JAPANESE LONG-TAILED FOWLS. 243

feathers were tied up liaving come undone, the feather got caught
in a door and was pulled out. There was some indication that
the feather was ceasing its growth, as, although the sheath was
long and soft, the harbs at the base were beginning to be slightly
downy.

The total lengt.h of the feather was 2 ft. IQi in., or 2 ft. 9^ in.
beyond the socket.

As this completes the history of the feathers of the first season,
I will here give a table showing the history of the feathers which
were pulled out, as it is the further history of these feathers
which constitutes the most impoi-tant result of my experiment.

Feathers of fiist adult plumage in Cock B pulled out.

Mght side of Tail. Ex^^L'^tl.

Small lateral coveit, growing Dec. 27.

Anterior lateral covert, nearly ceased Feb. 1 1 .

Posterior covert, ceased Feb. 1 7.

Anterior lateral covert, growing Mar. 2.

Anterior 1 st covert, growing Mar. 1 1 .

Central lectrix, growing Apr. 1 3

Left side of Tail.

Cential rectrix, growing Feb. 15.

Antei'ior central covei't, growing Mar. 1 0.

Anteiior 2nd covei't, nearly ceased May 27.

Posterior central covert, gi'owing J^ily 1 3.

Of these it is important to note that all but the last two
sprouted again almost immediately, and continued to gi'ow,
though not very f;ist, till the general moult. The last two did
not appear again until the month of Septembei-, that is to say in
the moulting-season. It is evident, thei'efore, that the reci'escence
of a feather is influenced by the sea,son of the year. After the
moult, that is to say in wintei' and spring, there is a general
tendency to feather-growth, and if a feather is pulled out duiing
this time a. successor at once begins to grow in its place. After
the end of April, at any late in the second year of the bii'd's life,
the activity of feather-growth slackens or ceases, and feathei'S
pulled out after this time are not replaced till the following
moulting- period .

Jtdij 3lst. Age 1 year 1 month 18 days. — Found today that
Cock A had begun to moult his wing-primaries : there wei'e thi-ee
or four new feathers growing, and one of them was already half
its full length. Cock B was in the same condition, but had not
shed so many feathers. The wing-secondaries were moulting also.

Ati,g. Ath. — Found one of the outermost shoi-t tail-coverts in
Cock A growing again, the old feather having been moulted.
This was the beginning of the moult in the tail-feathers.

244 MR. J. T. CUNNINGHAM ON [Mar. 17,

I believe there were 10 tail-coverts on each side in each cock, but
the two outermost on each side wei-e small and unimpoi'tant. It
will be seen fi'om the list given above that only eight of the
covei'ts in Cock B had been pulled ovit, so that eight feathers wei'e
left, apai't fi'om the I'ectrices, which had completed theii- giowth,
and could not grow any longei- till they wei-e moulted ; while in
Cock A all the feathers except one had completed their growth,
and must necessarily have been moulted befoie they could have
grown again,

Aug. llth. — The longest feather in tail of Cock B now was the
central rectrix of the left side, which 1 measured today and
found to be 1 ft. 8| in. long.

The old feather was pulled out on Feb. 15, so that its successor
had been growing nearly six months, and had grown at the rate
of little more than 3 inches per month.

Aug. 16th. — One of the dark-headed hens had moulted all the
rectrices as well as some of the tail-coverts, so that the i-ectiices
were shed almost simultaneously.

Aiog. 24:th. — Found that Cock A had shed another shoi't tail-
covert, one of the transition-feathers, and also three of the outer
rectrices : the latter were the 6th and 7th on i-ight side and 6th
on left, counting from the centre.

Cock B had shed none of the i'ectrices oi- old tail-coverts yet,
but this morning when I was stroking the feathers the 1st anterioi-
covert on the right side came out. The pi-edecessor of this had
been pulled out on March llth, and it was in full growth. It was
13f inches in length, and this was the growth of five months.
There was no reason to regard the loss of this f eathei- as connected
in any way with the moult, as growing feathei's ai'e not moulted.

The feathers growing in Cock B in place of those which were
pulled out in the spring were as follows : —

Three anterioi- covei'ts on right side.
One posterior ,, ,,

One anterior covert on left side.
Two cential rectrices.

Sept. 1st. — The outermost rectrix on each side in Cock B came
out when I was stroking the tail : they were loose and just about
to moult. This may be regarded as the beginning of the moult
of the tail, although one of the saddle- hackles and of the inter-
mediate feathers had been moulted occasionally for some time.

iSe])t. 2nd. — The 2nd posterior covert on right side in Cock B,
a full-grown feather, was moulted. It was 1 ft. 7| in, without
the imbedded quill, 1 ft. 8| in, in exti'eme length,

iSe2}t. 'S7'd. — 6th rectiix on I'ight side shed in Cock B ; it was
7 1 inches long without the quill,

Se2}t. 4:th. — 1st (nearest centre) posterior covert in Cock B
moulted ; it was 1 ft, 7| in, long. Also one of the outei' anterior
coverts in Cock A moulted,

Sept. 5th. — In Cock B three rectrices and one covert on left side

1903.] JAPANESE LONG-TAILED FOWLS. 245

came out when I stroked the feathers : these were all feathers of
completed growth, which came out in consequence of the moult.

In the light-coloured hen, nearly all the rectrices and some of the
coverts came out when I tried them. The two central rectrices
in the hen were longer than the others, slightly curved downwards
at the tip, and speckled with bi'own at the boideis : length
22'5 cm. (9 inches) without ([uill. The other rectrices were broader
and black, length 19'5 cm. (8 inches) without quill : the quill was
1'8 cm. in length.

Sept. ^th. — More tail-feathei's moulted from Cock B this
morning. The 2nd anteiior covei't on left side pulled out on
May 27th ha-d just begun to appear, so that renewed growth was
defei'red till the moulting-sea.son.

Sept. 1th. — Nearly all the rectrices and longer tail-coverts which
had completed their groAvth had now been moulted in Cock B.
I had pulled many of them out without violence when they were
loose. By looking at the ba«e of the featheis, I could see when
they were loose, because the tiansparent hollow quill was pushed
out and was visible Iteyond the socket.

In Cock A only four of the rectrices and none of the longer
covei'ts have been moulted spontaneously.

The left central rectiix in Cock B, which was still growing,
measured today 2 ft. Oj in., or had grown not quite 4 inches in
foui' weeks.

Sept. \9>th. Age 1 year 2 months. — Today, when taking up
Cock A to examine him, I accidentally trod upon the gi'owing
feather the predecessor of which had been pulled ovit on Feb. 10th.
It was one of the anterior coveits on the left side, possibly the
first. The feathei- bioke off at the top of the sheath, the latter
I'emaining in the socket. It was nearly 18 inches long (44"5 cm.),
and would probably continue to grow from the same base.

This cock had still six long feathers not moulted, namely : the
central rectrices, the 2nd pair of rectiices, and two posterior
coverts.

In Cock B the fourth anterior covert on the right side, the
outer of the three growing again, was coming to an end of its
growth, showing downy barbs at the base, and it would be curious
to see if it were moulted.

This fact is worthy of particulai- notice, for it proves that a
feather which rephices one pulled out before the moidting-season
does not necessarily continue to grow through that season, and
it I'aises the question whethei' feathers which are growing when
the moulting-season begins, necessarily grow longei' or for a
longer time than those which are moulted in the natural way.
This question is answered to some extent by the subsequent
observations.

At the end of September I cea«ed to i-eside at Penzance and
came to London, and having obtained permission through my
friend Mr. F. E. Beddard, F.R.S., to deposit the fowls at the
Society's Gardens, 1 took them there on Oct. 1st. They were

246 MR. J. T, CUNNINGHAM ON [Mar. 17,

accommodated in two of the pheasants' aviaries. I brought them
from Cornwall in hampers in a night train, with the tails tied up
in papers, and none of the feathers were injured. While they
were at the Gardens I kept the tail of B tied up, while that of A,
in which there wei-e no long gi'owing-feathers, was left free. Both
the bii'ds were left at liberty. During the fortnight after their
ai-rival, all the remaining old feathers in both tails were shed, and
the moult was thus completed, while in B there were left the seven
long feathers which had grown in place of those pulled out earlier
in the year.

This observation disposes, at any I'ate for these specimens, of the
supposition that the tail-feathers of the breed are not moulted
when left to themselves.

Oct. 11th. — Took away Cock B and the light-colovired hen from
the Gardens and placed them in a private fowl-house. The left
central rectrix measured 2 ft. 5g in., showing a growth of about
5 inches in five weeks. I fixed up a shoi't perch on a stand, and
at first allowed both birds to fly up to it in the afternoon, then
later in the evening tied the cock to the pei-ch and unfastened its
tail till the morning.

Oct. 22nd. — I intentionally pulled out the fourth of the anterior
covei"ts on the right side which had stopped growing, in order to
make the new feather grow again. Thus there were left only six
long feathers, all growing.

Oct. 2'dth. — In the morning found the Cock B hanging from the
perch by the leg which was attached by the piece of tape.
Generally, if he tried to fly off when he was tied, he was able to get
back again, but this time the tape slipped down the vertical sup-
port, and he was unable to do so. In his struggles he had injured
the longest feather, the left central rectrix, at the base, and it
broke oW near the top of the sheath. The detached part was
2 ft. 5-|- in. long. The root of the feathei- Avas uninjured, and the
basal part continued to grow.

Nov. 2nd. — Measured the central rectrix of the right side, which
was now the longest feather: it was 1 ft. 11 in. in length.

Nov. 15th. — Examined Cock A at the Zoological Gardens. The
rectrices were nearly full-grown, except the two central pairs
which were shed last, and which wei'e only a few inches in length.
The covei'ts wei-e only a few inches in length, and were not likely
to reach the ground for some weeks.

Nov. 17th. — This evening when I removed the paper from
Cock B, I found that another of the long feathers had come out.
The bird had been free all day, and must have pulled out the feather
by catching it in some projection when wallowing in the eaith, or
must have accidentally pulled it out when pi'eening himself. It
was the second anterior covei't of the right side, and was 1 ft. 9^ in.
long (53'8 cm.).

There were now only four long gi-owing-feathers besides the
left central rectrix which was not lost but broken off.

Bee. Qth. — I brought away Cock A and the other three hens

1903.] JAPAXTSE LONfi-TAILED FOWLS. 247

from the Gardens. The new featliers in A hail grown no faster
than the new feathers in B, in fact they were not quite so long.
The longest feather in B hnng aliout 2 ft. 8 in. from the perch,
and all four of thellong feathers were still growing. In Cock A,
none of the featliers touclied the groimd yet.

Dec. 9t/i. — Another of the featheis in Cock B pidled out today,
namely, the central anterioi' covert on the left side. The loss was
due to accident when the bird was free, not to manipulation.

Since this date I had kept Cock B tied up nearly every day in
a sepai'ate house, to prevent any more accidents. The featheis
that weie pulled out were all sprouting again, and there were three
long feathers still growing, namely, the right central lectrix, one
of the posterior coveits on the light side, and the third anterior
covert on the same side. The fiist two were about the same length,
that is about 2 ft. 6 in.

In Cock A, the 3rd pair of rectrices had ceased to grow, and
were a little longer than the outer lectrices. The 2nd pair of
rectrices were still growing, and were not much longer than the
3rd pair. The cential pair weie glowing, and were only al)out
4 inches long. The tail-coverts were growing, but still very short.

In Cock B, the 2nd paii- of rectrices were growing, and about
9 inches long ; the 3rd lectrix on the I'ight side had ceased to
grow, and the basal sheath was peeling oS", the coi-responding
feather on the left side was still gi'owing.

Dec. 21st. — The third anterior covert on right side, which was
pulled out last spiing and grew again, appeai-ed to have almost
come to the end of its giowth : the basal barbs were downy, and
the sheath at the base had almost all peeled off. I had not pulled
it out.

Histoiy of Feathers of Second Growth in Cock B,
to Dec. 28, 1902.

Bight side of Tail.

1st Anterior Covert : the preceding feather was accidentally
pulled out while growing on March 11th, when it was 1 ft.
8 in. long. It was accidentally pulled out a second time
while growing on August 24th, when it was 1 ft. 1^ in. long.
A successor immediately sprouted in its place.

4th Antei'ior Covert : stopped growing, and formed a complete
quill a little before October 22nd, on which date it was pur-
posely pulled out, and its successor spi'outed in its place : it
was not measured. Originally pulled out December 27th,
1901, when it was 1 ft. 2 in. long.

2nd Anterior Coveit : preceding feather pulled out March 2nd
when 1 ft. 44 in. long: accidentally pulled out again while
growing on November 17th (1 ft. 9g in. long), and its
successor sprouted in its place.

3rd Anterior Covert: predecessor pulled out February 11th, when

248 MR. J. T. CUNNINGHAM ON [Mar. 17,

1 ft. 2|- in. long : had almost entirely ceased to grow, but had
not been pulled out : it was a little over 2 feet long.

Posteiior Covert pulled out February 17th, 1902, when it was about

2 feet long. Now still growing, and about 2 ft. 6 in. long.
Central Rectrix : predecessor pulled out while growing on

April 13th, when 2 ft. 9 in. long : was still in full growth, and
about 2 ft. 6 in. long.

Left side of Tail.

Central Rectrix : predecessor pulled out while growing on
February 15th, when 2 ft. 41 in. long: was accidentally
broken ofl'on October 29th, but continued to grow from the
base, and the basal portion is now about 7 inches long. The
total length of the feather, if it had not been broken, would
have been 3 ft. Oj in.

1st Anterior Coveit : predecessor pulled out growing on
March 10th, when 1 ft. 9| in. long: accidentally pulled out
growing on December 9th, when 2 feet long. Its successor
immediately sprouted again, and was just showing.

There were thus in Cock B only three feathers left of the eight
which had begun to grow before the moulting-season. The two
longest of these were about 2 ft. 6 in. long, while the third was
shorter and had ceased growing.

The other feathei's which began to gi'ow after the moult were on
the whole longer than the feathers in Cock A ; and the difference
between the tails in the two birds, the one artificially ti'eated and
the other left to nature, was sufficiently striking.

Summary and Conclusions.

These observations extended only to the 2nd moult, or the first
shedding of the adult plumage.

They show that when the feathers are not pulled or artificially
treated in any way, cai-e being taken to protect them and prevent
them from being accidentally pulled out, the growth continues
till about the end of March, when it ceases, the quills of the feathers
are foimed in the normal way and the featheis are moulted nor-
mally in the following autumn.

On the other hand, when the feathers are pulled out in the
spring, successoi's immediately spi'out in their places, continue to
grow till the following season, when they go on gi'owing without
moulting, except in some cases when gi-owth may come to an end
in the moulting-season.

In the cock whose feathers were stimulated by pulling, growth
did not go on at a more rapid rate, but continued for a longer time
and produced a longer feather. Thus in Cock A, no growth took
place after April 1st, and the maximum length was 2 ft. 4^ in.;
while in Cock B, growth continued till July 13th, and the
maximum length was 2 ft. 9^ in.

1903. J JAPANESE LONG-TAILED FOWLS. 249

I think it is tolei-ably certain that the.se experiments explain
the method adopted ))y tlie Japanese in the production of tail-
feathers ot" extreme length, and that nuich that has hitherto l)een
mysterious and inexplicable in the matter is now explained. The
statement that the feathers do not moult or moult oidy once in
three years, or that the Japanese have a secret method of pre-
venting the moult, is explained if we assume that the Japanese
fancier strokes the feather as Mr. iS[)arks stated, and that he either
delilierately pulls a. feather out when it shows signs of diminishing
growth, or that the feather is automatically pulled out in the
process when its growth diminishes ; because, iis I have shown, the
attachment of the feather is very feeble when the growth is ceasing.
As a rule, when a feather is pulled out its successor immediately
spi'outs again, and its growth is aftei'wards continue*! regardless
of the moulting-season. The results of my experiments in fact
are in complete agreement with the statement of the matter
furnished l)y Mr. Sparks.

The long-tailed cock in its perfection, tlierefore, is neither a sport
nor a breed, but a, pi'oduct of artificial cultivation ; and the excessive
growth of the fea.thers is the I'esult of stimulation applied to the
individual. The most important part of the stimulation is not
the mere pulling of the feathei', but the extraction of it which
causes the growth of its successor.

On the other hand, the method of treatment is applied not to
any breed at random, but to a particular bleed wdiich includes
seveial varieties of coloui-, and apparently two vaiieties of comb.
It can scaicely be supposed that the same treatment applied to
another breed would produce results at all compaiable, and we
therefore may conclude that in this special breed there is a special
and extraordinary tendency to growth in the tail-feathers and
saddle-hackles. The congenital peculiaiity is evidenced in the
case of Cock A in my experiment. The question therefore arises,
whether this congenital peculiaiity ha.s been developed entirely by
spontaneous variations and selection, or whether it has been
influenced l)y the excessive giowth artificially induced in every
generation. It is proba1)le enough that in most cases a cock which
showed the most rapid and most prolonged giowth was used for
breeding, but it is l)y no means certain that the cocks so used
were nevei- subjected to the artificial ti-eatment. The proliabilities
are rather the other way, that specimens which had been found
to respond to the treatment were used for breeding. It is a
significant fact that this is the only In-eed of long-tailed fowls in
existence, and that the method of treatment applied to it is so
elaboi'ate and so absolutely artificial, re(|uiring daily attention for
months and years. If a similar result could have been obtained
by selection alone, it is difiicult to understand why jioultry fanciers
in some part of the world have not made the disco\"ery.

The results of this investigation are remarkably in agreement
with the theory advocated in my l>ook ' Sexual Dimorphism in the
Animal Kingdom.' In that book 1 pointed out that wherever

250 SIR CHARLES ELIOT ON NUDIBRANCHS [Mar. 17,

hypertrophy of organs or parts occurred in one sex only, the parts
affected were subjected by the habits of the animal, as known by
observation, to special irritation and stimulation.

Postscript, added April 25th. — On March 9th, as the left central
rectrix in Cock B seemed to be ceasing to grow, I tried to pull it
out, but it broke off at the top of the sheath. The total length
from the surface of the skin was 4 feet. The right central rectrix
in the same bird on March 16th was 3 ft. 4 in. long and still
growing. At present it shows no sign of cessation of growth,
and has now been growing for more than a year. The appearance
of the two birds at the present date is shown in the two photo-
graphs (text-figs. 41, 42, pp. 240, 241).

2. On some Nudibranchs from East Africa and Zanzibar.
Part lU By Sir G. Eliot, K.C.M.G., H.M. Commis-
sioner for the East Africa Protectorate, F.Z.S.

[Received February 20, 1903.]

Cbratophyllidia africana, gen. et sp. nov.

One specimen from near Wasin, E. Africa, in 9 fathoms.

The living animal was described by Mr. Crossland, who dredged
it, as of a light greenish-yellow colour on the upper surface, but
with the foot, branchiee, and under side of mantle, white. The
back was very hard and smooth, but its most remarkable character-
istic was the presence of a number of papillae, consisting of round
or pear-shaped bodies set on stalks. The stalks as well as the
base and tip of these globes were white, but the middle part was
black, owing to a dense aggregation of black spots, which, however,
can be seen to be separate under a lens. The globes were quite
soft and the stalks flexible ; they shook Avhen the animal was
moved, but were not observed to execute any spontaneous move-
ments. The mantle-edge was wavy.

The alcoholic specimen is of a uniform pale lemon-yellow, the
black bands of the globes being, however, preserved. The breadth
across the middle of the back is r9 centim. Unfortunately the
animal is contracted almost into a circle, but apparently the
length, when stretched out, must have been about 2*2 centim.
The consistency of the body is like hard wax, and fragments of
the mantle, which is ample, could easily be detached with the
forceps. The whole dorsal surface is a mass of closely packed
spicules. It bears about a hundred of the stalked globes. They
are of very varying size ; many ai'e quite minute, but the largest
is about 3 millim. high including the stalk, and about 2 millim.
across the ball, which is quite soft and can easily be pressed flat.
They are distributed over the whole of the back irregularly, and
not in any pattern, but are perhaps thickest round the mantle-
edge, including the space in front of the I'hinophores. Both the

1 For Part I. see P. Z. S. 1902, vol. ii. p. 62.

1903.] FROM EAST AFRICA AND ZANZIBAR. 251

iliinophore-pockets aiul tlie anal papilla project ; the edges are
siiiooth. In the preserv'ed specimen the rhinophores are grey.

Tlie Inaiichiie are arranged in a circnit interrnpted only hy the
head and genital p.apilla. They vary in size, hut thougii in places
long and short branchia' seem to alternate, this cannot he said to
be the general ride.

Over the mouth are two tentacles each al)out l*;") millim. long,
and 1 millim. broad at the base. They are not directt^l side\va}'s
but straight forwai'd, and being set close together so that the
division is not visible, they appeal- to form a soi-t of head. They
aie united at their bases. The mouth is lai-ger than is usual in
this Older, and though it is suctorial is hardly poiiform. Though
the animal was dissected only three months after capture, the
internal organs were already much dried and shrivelled, the spirit
having appaiently been unable to penetiate the hard integument.
It was clear, however, that the Imccal organs are of the type of
Phyllidiopsis rathei- than Phyllidia. The buccal opening led into
a sausage-shaped tube about G millim. long and 2 millim. broad,
with muscular walls ti'ansversely stripeil. This passed into a long,
narrow, coiled tube, which preser\ed the same calibre until it
dilated into the stomach. Two ample glands (salixary ?) entered
the larger part of the tube on either side, but were not in any
way fused with it. The liver was large and undivided behind.

The central nervous system was enclosed in a thin capsule and
somewliat concentrated, the cei'ebi-al and pleuial ganglia being
hardly distinguishable and the pedal ganglia lying beneath them.
The eyes were large, black, and distinct. The genital mass was
much hardened, but the two spei-matotheca?, one white and empty
and the other black and full, weie quite distinct. It was impossible
to ascertain whether the glans was armed with hooks and whether
the folds on the dorsal wall of the pericardium (sometimes called
the pericardial gill) were present, but it is higlily probable that
the species possesses these family characteristics.

In virtue of its buccal apparatus this animal belongs to Bergh's
genus Phyllidiopsis, although the tentacles are not attached
through their whole length and are I'ather larger than is usual in
the Phyllidiadje. It is remarkal)le that the genus PhylUdiojisis
contains one species. Ph. papilligera, which has also black papillae
on the liack. To me, the presence of these dorsal papillfe seems a
peculiarity sufficiently marked to merit generic rank. If Echino-
doris is a genus, why should not Phyllidiad;e which have the same
peculiarity enjoy the same distinction ? I would propose to call
the genus Ceratophyllidia, and its characters will be : — Back
studded, ivith papilla' ; buccal apparatus in the knoirn species
similar to thai o/" Phyllidiopsis.

Pleurophyllidiella horatii, gen. et sp. nov.

One specimen from Wasin, East Africa. Mr. Crossland, who
captured it, gives the following notes on the living animal : —
"Three inches long. Mantle edyed with light salmon-colour : it

252 SIR CHARLES ELIOT ON NUDIBRANCHS [Mar. 17,

ends in two ridges near rhinophores, similarly edged ; front edge
of velum similarly edged. Back grey, mottled with a darker
shade, the top of the numerous longitudinal ridges being sprinkled
with clear black spots. Underside white. Gills also white.
Rhinophoies longitudinally lamellated, grey in colour, and can be
contracted or reti'acted, though the pockets do not seem very
definite or complete."

The somewhat bent alcoholic specimen measures 3 centim. fi^om
head to tail, but woidd be at least "5 longer if it were stiaightened
out. The breadth of the back is 1"5 centim., of the foot "6. The
foot is long and narrow, pointed behind and truncate in front,
the corners not projecting.

No caruncle or nuchal papillae ai-e visible, but it is somewhat
difficult to reconcile the head-parts of the preserved specimen with
the desciiption quoted above. According to a rough drawing-
made from the living animal, it would seem that the mantle-edge
passes between the ihinophores and forms a sort of velum in front
of them; but in the preserved specimen it appears to lie behind
them as in an ordinary Fleurophyllidia, and not to pass through
them at all.

The salient character of the genus is that there are no branchiae
and no trace of a bianchial cleft. Lateial lamella? are, however,
piesent. They are situated exclusively on the under edge of the
mantle, and not on the sides of the body. They extend from
the head to the tail, and are about 30 in number on each side.
They are irregular in size and shape. Some terminate befoie
they reach the mantle-edge, and some r-un from the mantle-edge
only halfway to the body. The genital papilla is 4 millim. and the
A^ent 1'2 centim. from the anteiior end of the body. Cnidophores
are distinctly visible round the edge of the mantle.

The moiTth is ventral, and forms a lai'ge transverse slit, with
slight indications of a T-shape. The jaws are j^ellow, nairow, and
united so as to form a shape like a boat. They bear no denticles,
but thei'e aie a few irregular coarse indentations of the edge, due
apparently to its being jagged by use. The radula consists of
about 30 rows, the formula for each of which is abovit 50 + 1 -|- 50.
The central tooth consists of a squaiish basal plate with a long
cusp, which bears about 10 denticles on each of its sides. The first
lateral is laiger than the others and resembles the cential tooth,
except that the denticles are only on the internal side. The
second and third laterals ai^e also denticulate ; the rest appear to
be simply hamate.

This form appears to be intermediate between Pletvroleura,
which has neither- branchial clefts nor side lamellae, arrd Pleuro-
phyllidia, which has both. I have indicated its afiinities by the
name Pleurophyllidiella.

B^OLIDIA MAJOR, 11. sp.

(Bergh, in Sempei-'s Reisen im Arch. Philipp., Malac. Untersuch.
vol. iii. pp. 778-780, 1880.)

1903.] FROM EAST AFRICA AXD ZANZIBAR. 253

One specimen from Chuaka, E. coast of Zanzibar, nnder a stone
between tides.

The living animal \va.s about 4 centiuietres long. The l)oily and
appendages were of a uniform greyisli white, with spots of a dull
opaque white. The whole animal closely resembled a kind of
detaclial^le sea-anemone which is very common at Chuaka, and
appears to be sometimes almost free-swimming.

The alcoholic specimen is 3 centim. long and 1 centim. l)roa(l
at the widest part, including the cerata. The foot is moderately
broad, and has fairly long tentacular expansions in fiont ; but
its most remarkable character is the size and distinctness of the
anterior gi-oove, which measures 2 millim. across. The upper lip
is separated into two parts by a deep cleft. Tlie oral tentacles are
large and very thick. The rhinophores are shorter and studded
with minute knobs, which, in the preserved specimen at any rate,
appear not to l)e set in rings. The cerata are much flattened and
almost leif-like, and the hepatic diverticula within them are
ramided. They b.^gin at the anterior end of the large peri-
cardial prominence, and are arranged in about 17 groups on each
side, each containing about 10 cerata. There are vexy distinct
gaps between the anteiior groups, and a l)road bare space runs
down the middle of the back, but towards the end of the boily the
cerata are huddled together and continue until the extreme tip,
there being no tail. The outermost ceiuta of all the rows are
smaller, and the inner considerably larger, but at the base of the
innermost are frequently quite small, some hardly larger than
tubercles. The genital oritice is below the first group of cerata,
and the lateral vent behind the second.

The specimen was only partially dissected. The jaws are very
large, colourless and transparent, with a perfectly smooth edge.
The radula consists of 32 pectinate teeth, very similar to those
df B. mcebii (see Bergh, /. c. pi. IxxLx. fig. 16), with striations
under each denticle. They are, however, very much broatler, the
widest measuring 2 millim,, and the denticles are more irregular
in shape, being pi-obably worn by use. There are about 150 of
them on the broader teeth. The three or fom- central denticles
are generally, but not always, smaller than the others.

This specimen is clearly referable to Bergh's genus BieoUdia,
and the difference between it and the type is mainly one of size,
B. mtddi being only 8 millim. long. The similarity of habitat
makes one thirdv that this may be merely a full-grown individual
of the same species ; and we know so little of the variations which
the radula and arrangement of cerata may present in .'Eolids at
the different periods of their growth, that I am not prepared to
reject this hypothesis. Still, the single specimen examined by
Bergh appears to have been sexually mature, and this being so
the two animals each present peculiarities amounting to specific
diflferences: — (1) In B. mcehii the tentacles are said to' be " jibge-
plattet fingerformig" ; the cerata begin behind the rhinophores and
are set in rows: in B. major the tentacles are stout and round;

254 SIR CHARLES ELIOT ON NUDIBRANCHS [Mar. 17,

the cerata begin further back and are set in groups. (2) In
B. mcebii the reproductive orifice is under the third row of cerata,
and the vent between the sixth and seventh rows, ahiiost dorsal :
in B. major the reproductive orifice is under the fii'st group of
cerata, and the lateral vent after the second. (3) In ^. inajor
the basal part of the teeth is proportionally narrower than in
B. 7ncebii, and the denticles are more irregular.

Var. ORNATA.

One specimen captured at Chuaka, May 1902, seemed to be a
typical Bceolidia major, except for a somewhat more ornate
coloration. The ground-colour was of a yellowish-white with a
yellowish-brown pattern, consisting of a series of irregularly shaped
lozenges containing white spots, down the middle of the back as
in B. major. The oi'al tentacles were white with green stripes.
The cerata were white with yellow tips, below which was a bright
blue band.

CeRBERILLA AFRICAN a, n. Sp.

(Bergh, Journ. Mus. Godef. Heft ii. 1873, & Heft vi. 1874;
Beitr. zur Kenntniss der Aeolidiaden, Theil ii. ; Semper's Reisen
im Arch. Philipp., Malac. Untersuch. iii. p. 879.)

One specimen from thereof Jembiani, Zanzibar, 3*5 centim. long
and 1*3 broad in life. The back was almost entirely covered with
cerata, so that the body-colour was hardly visible. Most of the
cerata were very dark green with a blight yellow ling, but the
innermost were white with bluish tips, with only a ring of dark
green. The oral tentacles were dark blue, with green bases. The
rhinophores had four bands of colour, which were, starting from
the base, greenish brown, white, blue, white. On the head were
two yellow lines, extending from the oial tentacles to the rhino-
phores, and the margins of the head and foot were also of a bright
light yellow.

The animal was stoutly built. The foot projected considerably
beyond the body on either side. Its anterior angles were expanded
into long tentaculai' processes. The head had also two lateral
expansions, from which projected at right angles the very large
and conspicuous oral tentacles. In life the rhinophores were
quite simple and fairly long. In the alcoholic specimen they were
contracted and somewhat wrinkled. The numerous and thick-set
cerata wei-e disposed on peduncles. There was a bare triangular
patch behind the rhinophore, and a narrow bare space down the
middle of the back, but the cerata folded over the latter so that
neither it nor the transverse bare areas were visible. There were
about twenty transverse rows of cerata. The first row at the side
of the rhinophores consisted of about 1 cerata, much smaller than
the rest. After the third row Avas a distinct gap, and a smaller
gap after the fourth. After that the rows were so close together
that they could not be distinguished superficially. The innermost
cerata were larger than the others, and sometimes bifid : small
cerata extended almost to the end of the very short tail.

1903.] FROM EAST AFRICA AXD ZANZIBAR. 255

The jaws were large, witli a smooth cutting-edge. The radula
consisted of a single seiies of 13 tiunspai-ent yellowish teeth of the
shape usual in the genus, viz., pectiniforin, with large irregular
denticles and small accessory denticles. It was sometimes hard
to decide whether the I'ather small denticles should he considered
main or accessory ; Imt tlie average number of main denticles on
each tooth was 10, and the largest number (in one case only) 13.
The central nervous system was somewhat concentrated. The
specimen was only jjartly dissected.

A new species must, I think, be provisionally created for this
animal, though the discovery of intermediate foi-ms may perhaps
render its retention unnecessary. In some ways it is itself a con-
necting-link between C lonc/icirrha {\nd C. annulata, for the former
is Siiid to have 7, and the latter 5 denticles on each side of its
teeth, whereas C. africana has 4, 5, 6, oi- 7 indifferently. It can
hai'dly be C. annulata, for the difference in colour is too great,
and besides thei-e is much less bare space on the back. Neither
can it be C. longicirrha, because (1) the coloiution, though similar,
is still distinct ; (2) C. longicirrha has the back bare up to the 7th
row of cei'ata, and some of the cei'ata ai'e veiy long, which is not
the case here ; (3) the rhinophores are not perfoliate.

This last point is of some importance for the chai'acterisation
of the genus. In the present animal the i-hinophoi-es were
undoubtedly quite simple in life, and in alcohol they ai-e wrinkled,
though it is still possible to see that they ai-e not really peii'oliate.
In C. longicirrha, Bei'gh says the perfoliations ai-e 14 or 15 in
number, and not deep. Of C. annidata he says that the i-hino-
phores have 12-14 well-marked perfoliations, and that Gari'ett has
wrongly represented them as simple. But in Semper's ' Reisen,'
xvii. he states that C. annulata var. affinis has simple i-hinophores,
and gives as a generic charactei* : " Die Rhinophoi-ien scheinen
nicht pei'foliirt zu sein." I have not access to part ix. of his
' Beiti-iige zur Kenntniss der Aeolidiaden,' which j^erhaps explains
the matter ; but it looks as if the I'hinophores are simple, but have
a tendency to simulate perfoliations when preserved.

Pteraeolidia semperi.

(Bergh, Beitr. zur Kennt. der Aeolidiaden, iii. p. 22, and in
Semper's Reisen, Malac. Untersuch. vol. i. p. 18 (1870); under
Flahelliiia.)

Four specimens, which seem probably referable to this species,
were di-edged fi-om 3 fathoms near Chuaka in July 1 901 . The body
is veiy long, narrow, and vermiform, the lai-gest individual being
5*5 centim. long and only 3 millim. broad. The ground-colour of
the body in the living animal is bi'own, with opaque mai-kings of
veiy light green on the sides and back. The ceiuta are also dark
brown, with numei'ous thin lines of the same green. The top of
the head and the ends of the oi-al tentacles ai-e opaque yellowish
\vhite. The lowei- part of the tentacles bi'own, with three i-ings of

Proc. Zool. Soc— 1903, Vol. I. No. XVII. 17

256 SIR CHARLES ELIOT ON NUDIBRANCHS [Mar. 17^

vivid crimson-lake. The body of the rhinophoi'es is brown and
inconspicuous, but the tips ai-e whitish and have a crimson ring.
Although the colour of the animal when analysed is as desci-ibed,
the general effect in most lights is that it is purple with a silvery
glaze.

The specimens were active in their movements, and in particular
the long oral tentacles were very mobile. The anterior margin of
the foot was produced into two deeply-grooved processes, and its
sides projected in two laminse along the whole length of the body.
The cerata, which were not at all caducous, were set on fan-like
pr-ojections of the dorsal margin, from 13 to 20 in number, on
each of which were from 10 to 20 cerata. The largest individual
appeared to have about 640 cerata in all. None of the cei^ata were
large, but they became more crowded and smaller towards the end
of the body, where they extended to the tip of the tail. In three
of the preserved specimens there was a very distinct raised border
on each side of the back, connecting the fan-like projections, and
almost resembling the mantle-edge of Dorids ; but in the fourth
this I'emarkable character was not visible. The oral tentacles wei-e
ver};' large and long ; the rhinophores, which stand between the
first clumps of cerata, were small and lamellated, except the tips,
which were nari-ow and smooth. In the individual dissected the
radula consisted of a single series of 18 yellowish teeth. The
separate teeth were much as represented in Bergh's plate (I. c.
pi. iii. figs. 5, 6, 7), but somewhat more regular in shape, having
nine denticles very symmetrically arranged on each side of a
central cusp, but not taking the foi-m of indentations of the cusp
itself. The cutting-edge of the jaw bore a row of fine but very
distinct denticles, at the side of which were several I'ows of less
distinct accessory denticles. No spine or other armature was
discovered in the I'epioductive system.

These specimens seem to clearly belong to Bergh's genus Pteraeo-
lidia. There are some small differences (such as the shape of the
rhinophores and of the teeth and the lateral ridges) betAveen them
and his description and plates of Pteraeolidia ( FlabeUina) semperi^
but the ridges wei-e absent in one specimen and the other characters
were slight. It is possible, however, that a comparison of the living
animals might shoAV a specific difierence.

Ercolania zanzibarica, n. sp.

(For the genua Ercolania seeTrinchese, " Un nuovo genere della
fam. degli Eolididei" Ann. del Miis. di Stor. Nat. di Genova, ii.
1872; id. " Aeolidsedel Porto di Genova"; Bergh, Beitr. z. Kennt.
der Aeolidiaden, v. pp. 9-18, Wien, 1878; Yayssiere, Rechei'ches
sur Moll. Opisthobranches, Marseilles, 1888, pp. 121-128.)

Two specimens captured at Chuaka, East Coast of Zanzibai-,
in February 1901. The living animal was rather more than
2 centimetres long, very fragile and delicate, transparent and of a
uniform bright gi-een, the hepatic diverticula in the cerata not

1903.] FROM EAST AFRICA AND ZANZIBAR. 257

liaving been distinguished by any diflerence or intensity of colour.
It exactly resenililed a piece of the x-acemose seaweed {Caulerpa)
on Avhich it was found.

The length of the alcoholic specimens is 2 centim.; the extreme
breadth of the l)ack with cerata 8 millim., and of the foot 2-3 millim.
The rhinophoretj are long and distinctly canaliculate. There are
no oral tentacles, but two lobes ovei- the mouth. Behind the
rhinophores are two very distinct black eyes. The cerata are
clu1>shaped as in Galvina, of varying size, the largest inside. On
each side of the back are four clumps of about nine cerata each,
and there is a thick bunch on the tail, which, however, projects a
considerable distance behind the last cerata. Down the centre of
the back is a broad bare space, in the anterior portion of which
is the very large, elongated (not oval) pericardial prominence. In
front of this and^ fused with it is the vent, a large and conspicuous
tube. The foot is rounded in fi'ont.

I dissected one specimen, but was unable to obtain a clear view
of either the central nervous system or the reproductive organs.
The latter, as usual in this family, Avere extremely complicated,
both the prostate and albumen-gland appearing to be extensively
ramified. The verge was armed with a small spine. The hepatic
diverticula in the cerata, being colourless, were not easy to dis-
tinguish, but appear not to be much ramified and to resemble the
figure of those in Ercolania siottii in Trinchese, pi. ix. fig. 2.

The mouth-parts, buccal muscles, radula, &c. are of the usual
ascoglossan type. The teeth are not unlike those oi Ercolania
viriclh (v. Bergh, I. c. pi. xii. figs. 3 & 4), but the doi'sal surface
is a simple curve and does not show any depression. The upper
portion of the radula contains 6 teeth, the lower 27, the last
members being arranged in a spiral like that depicted in Trinchese's
plate of B. siottii (pi. x A. fig. 1), from which it may be concluded
that the individual is young.

As the specimen presents all the characters of the genus Frco-
lania, I describe it under that name, but I feel very doubtful if
the genus is valid. The only characters which difierentiate it
from Stiliger, viz. that the rhinophores are canaliculate and the
pericardial prominence elongate and not oval, are surely very
slight. Yayssi^re [l. c. p. 122) referred to the genus a species
(fauerea) with entire rhinophores, which is probably in any case
a Stiliger.

The animal is not likely to be specifically identical with
E. viridis Bergh, for the coloration is not really the same, the size
is much larger, and the shape of the teeth somewhat diflferent.

17*

258 MR. W. p. PYCRAFT ON THE - [Mar. 17,

3. Gontribiitions to the Osteology of Birds.
Part VI. Cuculiformes} By W. P. Pyckaft, F.Z.S., A.L.S.

- [Received March 17, 1903.]
(Plate XXII.' & Text-figures 43-48.)

COMXENTS.

i. Introductory Remarks, p. 258. I vii. The Pelvic Girdle, p. 280.

ii. The Skull of the Adult, p. 258. viii. The Pectoral Liinb, p. 283.

iii. The Skull of the Nestling, p. 268. ix. The Pelvic Limb, p. 284.
iv. The Vertebral Column, p. 271. x. Summary, p. 286.

V. The Ribs, p. 274. xi. Key to the Osteology of the
vi. The Sternum and Shoulder-girdle, Cuculiformes, p. 288.

p. 275. I xii. Explanation of the Plate, p. 291.

i. Introductory Remarks.

Although numerous contiibutions to the osteology of the
Cuculiformes have been made during the last quarter of a
century, the present paper claims to be the most comprehensive
survey of the group from this point of view which has yet been
made. Nevertheless, much remains to be done before our know-
ledge is anything like complete on this subject. How much, may
be gathered from the fact that out of 45 genei'ally recognised
genera of the Family Cuculidfe, only 20 are represented in the
osteological collections of the British Museum, and this appears to
be the lai-gest collection of the group extant. Of the 6 genera of
the Family Musophagidae, only 2 are represented in our series.

ii. The Skull of the Adult.

The skull of the Cuculiformes recalls, on the one hand — through
the Cuculi — the skulls of certain Coraciiformes, e. g. Coraciidce
and Buccones ; and on the other — through the Musophagi — the
skull of the aberrant Gallifoi'm OpistJiocomus.

Nevertheless the Cuculiform skull may at all times be readily
distinguished, and by the following chai-acters : —

The pterygoid is always fi'ee, and aiticulates with the palatine
— after the fusion of the hemipterygoid — by a more or less
obliquely transverse joint ; the lachrymal is always present and
free ; the antorbital plate (prefrontal) is large, and generally (if
not always) bears an os uncinatum ; the vomer is vestigial or
wanting ; the palate is indirectly desmognathous ; basipterygoid
processes are wanting ; and the postorbital processes are small and
inconspicuous, never extending down to the level of the quadrato-
jugal bar.

The skull of the Musophagi may always be readily distinguished
from that of the Cuculi by the fact that the mesial borders of

1 For Part V., see P. Z. S. 1902, vol. i. p. 277.
' For explanation of the Plate, see p. 291.

P.Z.S.I903,vol.I.Pl.XXII.

'■° s\ T 'as 8

m

na 7.

als fr

pro
me

ooc.

H Gronvold.del Photogravure by Bale iDaniel?, r-

OSTEOLOGY OF THE CUCULIFORMES

1903.] OSTEOLOGY OF THE CUCULIFORMES. 259

the palatines ai-e cut away so as to expose the pai'as^ihenoidal
rostrum. _

The Occipital Region.

The plane of the occipital foramen slopes ol)liquely l)ackwai'ds ;
the oiifice is either i-eniform or cordiform. In the obliquity of
this foramen the Cuculifoim skull difters conspicuously from many
Coraciifoi'in skulls whei'ein the foramen looks directly do^\^l^vards,
its vei'tical axis lying- parallel with the long axis of the skull.
The supr a-f or amincd ridge \iihu.ti\\-i\c&\\Q(\, but can be indistinctly
ti-aced I'unning downwards, on eithei' side, into the processus alse
occipitalis.

Tlie lamhdoidal ridge in the Oucnli is generally well-defined and
bluntly conical in outline. In some genera, e. g. Crotophaga,
Centropus^ Gicira, Rhojyodijtes, Rhinococcyx. and Saurothera, its
extremities bifurcate, sending foi-ward an anterior band aci-oss
the inferior bordei- of the squamosal to terminate in a pointed
squamosal prominence ovei-hanging the quadrate ; and a posterior
bi'anch, which turns downwards and backwards to disapjDear on
the free edge of the latei'al occipital. "VVhei-e the lambdoidal
ridge is well-developed, the ci-anial I'oof, \ying immediately in
front, is more or less flattened, often, indeed, marked by deep
temporal fosspe.

In many genei^a the lambdoidal ridge is much less distinct.
This occurs where the ci'anial roof is sti-ongly arched so as to rise
above the ridge with a strong curve, and temporal fossae are
wanting or confined to the sides of the cranium, e. g. Coua^
Cacomantis, Chrysococcyx.

The supi-aoccipital i-egion is marked by strong muscular depres-
sions, which in the case of Coua, for example, take the form of
widely-separated oval scars, each surrounded by a swollen border ;
whilst in others, as in Eudynamys, these depi-essions are only
divided in the middle line by a narrow ridge, and are bounded
infei'O-laterally by slightly swollen bullse produced by the lateral
occipitals.

The latei'al occipital wing is j^roduced downwards on either side
into a small or medium-sized jyrocessus alee exoccipitalis inferior
(paroccipital process).

In the Musophagi the lambdoidal ridge is fairly well-defined in
Schizorhis, but only vei'v imperfectly so in Turacus.

The i-idge in Schizorhis is formed by the scooping-out, so to
speak, of the supi-aoccipital to aftbrd depressions for the insertion
of the cervical muscles. On either side, and inferior to these
depressions, are two rather conspicuoiis bulla? lodging the posterior
and horizontal semicii-culai- canals of the ear. The distal ends of
the lambdoidal ridge tei-minate before reaching the fi-ee edge of
the lateral occipital wing. The 2»'0cessus alee exoccipitcdis ivfeHor
is only feebly developed.

The lateral or exoccipital wings are not gi'eatly developed in
the Cuculiformes.

260 MR. W. p. PYCRAFT ON THE [Mar. 17,

The Cranicd Roof. — The cranial roof is never marked by supra-
orbital grooves.

The parietal region, in certain genera of Cuculi, is marked by-
deep temporal depressions (temporal "fossae "), which may meet in
the middle line so as to form a faint sagittal ci-est, e. g. Rhojjodytes,
Piaya, Rhinococcyx, and Rhcmiphococcyx ; or they may be divided
by a low broad ridge, as in Geococcyx, Taccocoiuc, Satf^rothera,
Exidynamys, Guira, Centropus, and Crotophaga. In Goua,
CoGcystes, Guculus, and Gacotnantis the temporal fossa is either
barely perceptible or confined entirely to the lateral aspect of the
cranium. Geococcyx, it should be remarked, presents an inter-
mediate stage in this character ; the width of the ridge dividing
the right and left fossae being extreme, so much so that they
barely pass beyond the latei'al aspect of the skull.

The interorbital region is moderately wide in all the Cuculi,
save in Saurothera, but even in this genus it is not conspicuously
narrow. In the middle line the region may be marked with a
distinct furrow as in Goua, or even by a low ridge as in Eudynamys.
The interorbital region is bounded anteriorl}- by the lachrymals ;
these do not develop outstanding supraorbital processes, and are
not conspicuous from this aspect of the skull. There is no trace
of the fusion of the fi'ontal and rtasal bones.

In the Musophagi the temporal fossae are not strongly marked,
and are confined to the lateral aspect of the skull. The parietal
region is more rounded than in the Cuculi ; and in Tttracus is
marked by a median furrow, so that the foi'm of the cerebral
hemispheres is distinctly indicated. In other I'espects this region
of the skull resembles that of the Cuculi.

The Base of the Skull.

The basitemporal plate, in the Cuculi, is flattened and triangular
in shape, and with the apex lying immediately beneath the
Eustachian apertures. Posterioily it is bounded by a pi'econdylar
fossa, which, though generally barely perceptible, in Eudynamys,
Gentrop'iis, and Geococcyx is fairly deep. The lateral angles of the
base are truncated, and, turning sharply downwards, ci'oss the
paroccipital notch to form the lower portion of the rim of the
mouth of the tympanic cavity. In some genera, e. g. Goua, Gidra,
these downwardly-directed angles are very prominent and simulate
mammillary processes.

The two sides of this tiiangular plate have fused with the
ossified connective tissue foiining the anterior wall of the recessus
tympanicus anterior. Consequently the Eustachian grooves have
been converted into tubes opening directly above the apex of the
triangle.

The parasphenoidal rostrum bears slight vestiges of basi-
pterygoid processes in Eudynamys only among the Cuculi.

The rostrum is relatively long, and curves slightly upwards.

In the Musophagi the basitemporal plate, like that of the

1903.] OSTEOLOGY OF THE CUCULIFORMES. 261

Cuculi, is triangular ; l)iit it tlifters from the Cuculine plate in that
it is somewhat swollen, a feature which is especially well-marked
in Schizorhis. The two sides of the triangle are produced hack-
w^ards over the paroccipital notch to join the lateral occipital wing
of the tyinpanic cavity : thus they come to project beyond the
base of the triangle, though they are less conspicuous tliau in
some of the Cuculi. There is a barely perceptible precondylar
fossa.

The Lateral Aspect of the Cranium.

The tympanic cavity is oval or oblong in form and of moderate
size. It is bounded in front by the quadrate, behind by the
lateral occipital tympanic wing, below by the basitemporal jilate,
and above by the squamosal prominence and by the head of the
quadrate.

Within the mouth of this cavity lie several smaller apertures.
The largest of these is that of the recessus tympanims anterior.
Immediately below this is the mouth of the Eustachian tube.
The foramen ovale and foi'amen rotundmn, three foramina of the
sinus petrosus, and the mouth of the posterior tympanic recess
open into the cavity by a common aperture — tliefenestral recess —
which is very small, and lies mesiad of the articular surface for
the otic head of the quadrate. The mouth of the posterior
tympanic i-ecess is completely concealed, and can only be made
out by cutting away its outer wall and passing a bristle through
from its cavity into the common apertui'e of the foi'amina in
question. In some birds, e. g. Falco, the posterior tympanic recess
communicates with the tympanic cavity by two apertures— one
caudad of the foramina ovale and rotundum, and lying within the
fenestra through which these are appi'oached ; and the other external
to this fenestra, separated therefrom by a bony column, and lying
immediately beneath the articulai- surface for the otic head of the
quadrate. This external aperture is wanting in the Cuculiformes.
Immediately above the head of the quadrate, and between the otic
and squamosal heads, will be found the aperture of the superior
tympanic recess, which is reduced in this group to extremely sinall
dimensions. The foramen of the 7th (facial nerve) opens also into
the tympanic cavity between the otic articular surface for the
quadrate and the lim of the anterioi- tympanic recess.

The squamosal and otic articulai- surfaces for the quadrate form
a dumbbell-shaped area immediately in front of the aperture of
the superior t}-mpanic cavity. The actual articidar suiiaces are
formed by the expanded ends of the dumbbell only. Where the
pneumatic area is ver}^ large, as in the Striges and Falconiformes
for example, it breaks thi-ough the connecting poi'tion of these two
articulai- sui-faces so that they liecome quite isolated.

The Squamosal Prominence.— The size of this prominence
depends largely ujton the development of the temporal fossa.
When this is deep the prominence is large ; but when shallow the
prominence is quite inconspicuous. Its anterior angle is produced

262 MR. w. p. PYCRAFT ON THE [Mar. 17,

forwards into a more or less well-developed spine, the p7-ocessus
zygomaticus squcmnosi. The base of this angle serves for the
articulation of the squamosal head of the quadrate. This zygo-
matic process is especially well-developed in Cuculus, Crotophaga,
and Geococcyx. It is small in Gidra and Goua, for example,
and in the Musophagi. The j^'^^ocessus articularis squamosi is
wanting.

The temporal fossse are never very deep, but they may extend
inwards to within a short distance of the middle line. The
variations in the extent of these fosste have already been dealt
with {supra, p. 260).

The trigeminal foramen in the Cuculi lies immediately above
the mouth of the anterior tympanic recess, on a level with, and in
front of, the otic ai-ticular surface for the quadrate. In Turacus
this foramen is divided by a median vertical bar of bone.

The orbito-nasal foi-amen (v^) appears to be definite only in
Geococcyx. In other Cucidinoi and Ihosophagioue the alisphenoidal
wall only partially ossifies, and consequently this and other
foramina aiound the optic foramen disappear in the dried skull.

The orbits are large, and roofed by moderately broad over-
hanging ledges formed by the frontals, and to a slight extent by
the lachrymals. The orbito-s2)henoid does not ossify.

The interorhital septttm in the Cuculi is genei-ally pierced by a
large fenestra, which attains its maximum in Dromococcyx and
Centropus. In Grotophaga, Goua, Piaya, Rhamphococcyx, Rhino-
coccyx, Rliopoclytes, and Zanclostomus, however, the fenestra is
greatly reduced.

In the Musophagi the fenestra is large in Turacits, small in
Schizmdiis. I have not had an opportunity of examining other
genei'a.

The oi-bit is bounded in front by the antorbital plate (p. 263).

The lachrymal in the Cuculi is moderately large, with subequal
supraorbital and descending processes. The form of this bone
and its relation to the frontal and nasal bones vary considerably
however. In Rhamphococcyx calorhynclms only, apparently, is it
attached merely to the nasals. The orbital process looks entirely
forwards, instead of upwards : that is to say, it forms a shield
for the fiont of the eye instead of an ovei-hanging ledge therefor ;
the descending pi-ocess is slender. The frontals rest on the inner
angles of the orbital process of the lachrymal on either side, and
appear, in the adult skull, to be cut ofi' from the nasals and pre-
maxillary process by a shai'ply-defined groove, which runs across
the skull fi'om the hinder edge of the orbital process up the
lachrymal on the one side to that of the other, and simulates a
nasal hinge. Rhinococcyx curvirostris appai'ently closely resembles
Rha'mphococcyx, but the lachrymal being completely fused with
the nasals and the frontal behind it, the precise relations of the
various regions cannot be made out.

In othei- Cuckoos, though exhibiting a distinct tendency to
form an anteiior shield rather than a jDent-house ledge, the

1903.] OSTEOLOGY OF THE CUCULIFORMES. 263

lachrymal extends backwards for some distance over the fiontals.
In Eadj/namys the fi-ontals ai-e deeply notched to receive the
orbital process of the lachiymal, which is more or less diamond-
shaped. In other genera the orbital process of the lachrvnial
may be semilunai' or rod-shaped. Whei'e the antorliital process
is very lai-ge, the descending pi'ocess of the lachiymal is always
more or less markedly degenei'ate. In Cuculus it has disappeai-ed
altogether, wliilst the orbitnl pi'ocess has fused with the frontal
and nasal bones. In Zanclostomus javanicus it is I'ednced to a,
mere vestige. The descending process is unusually large in
Crotopluiga, one of the genei'a, it will be i-emembered, with a small
antorbital pi'Ocess.

Geococcyx has the lai'gest lachiymal of all the Cuckoos. It is
nearly columnar in form, flattened antero-posteiiorly, and deeply
gi'ooved immediately above the quadrato-jugal bai'.

The lachrymal in the Musopliagi in its genei'al sha])e closely
resembles that of the Cuculi : the oi'bital process is, however,
inclined more upwards. The descending process, judging from
what obtains in Schizorhis and Jlnsophac/a , is faiily strongly
developed. In Schizorhis the fi-ee end of the descending process
is twisted in itself, i-ests upon the quadrato-jugal bar, and ai-ticu-
lates by its postero-inteiiial angle with the enormous ossicidmn
lachrT/mo-jxolatinum.

The lachrymo- nasal fossa varies considerably in size. Among
the Cuculi, it may be described as large in Eudynamys, Centropus,
Gidra, and Crotophaga. In Coua, Rhamj^hococcyx, and Geo-
coccyx, for example, it is extremely reduced by the encroachment
of the lachiymal.

The Ethmoidal Region. — The onesethmoid, both in the Cuculi
and the ]\Iusophagi, terminates abrupth' immediately in front of
the antorbital plate. The horizontal plate formed by the expan-
sion of its dorsal bordei- is but feebly developed.

The antorbital plate (prefrontal), which, with the lachiymal,
bounds the orbit in front, is, in some Cuculi, e. g. Cacontantis,
Centrojms, Coua, C'occystes, Cacidus, Eitdynamys, Geococcyx,
Fiaya, I'accocoua, and Saurothera, conspicuously large, quadrate
in form, and has its dorsal liordei- piei'ced by two foi'amina — an
inner for the passage of the olfactory nerve, and an outei' for the
orbito-nasal (v^).

In other foims, e. g. Gaira, Crotop)haga, the antorbital plate,
though large, has the external lateral and inferior borders
deeply excised. In such cases the orbito-nasal foramen described
above is repi-esented by a notch. Attached to the inferior border
of this plate is a vestigial ossiculum palatinum, which, however, is
generally wanting in dried skulls.

In the Musophagi the antorbital plate is greatly reduced,
forming but a triangular process projecting from the mesethinoid.
From its inferior border depends a well-developed ossicxdum
lachrymo -palatinum. In Schizorhis tiiis bone is relatively
enormous, projecting downwards to aiticulate by a strap-shaped

264 MR. w. p. pycRAFT ON THE [Mar. 17,

limb with the outer border of the palatine. Its further relations
with the lachrymal will be discussed presently.

The nasal septum is completely ossified, and the external nares
are in consequence imperforate. The ectethmoidal region pi-esents
several points for comment. By the ossification of the alinasal
wall, the form of the external nares is greatly changed in many of
the Cuculi. In consequence of this ossification, the space between
the premaxillary and descending processes of the nasal becomes
obliterated. Thus in Geococcyx the nostrils take the form of
an oval aperture lying just cephalad of the middle of the beak.
In RhampliocoGcyx this aperture is still further reduced, and
consists only of an elongated slit lying at the base of the beak
above the tomium. In Rhinococcyx the aperture is nearly circular
at the base of the beak, and bounded by a deep groove in front.
The alinasal ossification is, however, impei'fect, inasmuch as a
small round hole is left in the angle betAveen the premaxillary
and descending processes of the nasals. In Taccocoua this
superior foramen and the aperture of the nostril have blended.

In Coua the floor of the anterior ]'egion of the nasal fossa is
raised up into a sharp longitudinal I'idge.

In the Musophagi the foiiii of the external nostril diifers con-
spicuously from that of the Cuculi. It is situated slightly distad,
or proximad, of a line passing through the middle of the beak, and
is renifoi-m in shape. In Titracus it lies distad, in Schizorhis
proximad of the middle line. In both genera, just within the
mouth of the aperture, lies a well- developed concha vestibidum.
This is especially large in Schizorhis. Furthermore, the genus is
peculiar in that immediately distad of the concha the nasal septum
is pierced by a minute foramen.

The Cranial Cavity. — The metencej)halic fossa is somewhat
shallower in the Musophagi than in the Cuculi. The internal
auditory meatus in the Cuculi is sharply defined, strongly con-
trasting in this respect with the Musophagi, in which it is but
ill-defined. The vagus foramen, which lies caudad and ventrad of
the meatus, is small. The orbito-nasal forameoi (v^) pierces the
outer superior rim of the fossa near the outei- angle of the dorsum
selloi.

The cerebellar fossa is relatively smaller in the Cuculi than in
the Musophagi, and in both it is relatively smaller than in many
other groups, e. g. Falconiformes. In the Cuculi the floccular
fossa is very shallow, a.nd lodges a deep slit-like vertical
depression. In the Musophagi this fossa is reniform, deeper, and
lodges a deep and wide depression which gives the reniform
character to the whole.

The mesencephalic fossa is fairly sharply defined in both groups.
Among the Cuculi, this is especially noticeable in Geococcyx. Its
floor, in both groups now under consideration, is pierced by the
trigeminal foiumen. The diflference in position of this foramen is
worth noticing. In the Falconiformes it lies, together with the
orbito-nasal (v'), in a pit carved out of the superior border of

1903.] OSTEOLOGY OF THE CUCULIFORMES, 265

the metenceplinlic fossa, midway between the sella turcica and the
floccular fossa, and is overhung by a sharp ledge formed by the
inferior border of the metence'phalic fossa.

The pitidtarj/ fossa is large, deep, and tubular. The hinder
border of the fossa— the dorsum seZ^re— forms a tumid lip.
Anteriorly the fossa is bounded by the 2)re-piti(.itan/ ridge, which
forms a broad triangular optic platform. This platform is carried
far forward into the interorbital septum, rising gently in its
course, and then turning abi'uptly backwards to terminate at the
p7-e-optic ridge, which may be traced into the tentorial ridge.
The form of the optic platform contrasts strongly with that in
some otlier types, e. g. Accipitres. The tentorial ridge in the
region of the pre-optic platform is but feebly developed. On the
parietal wall, however, it becomes tolerably distinct, especially in
Geococc]/x and the Musophagi. In these, on its Avay to the median
falx it meets, near the crest of the epiotic, the sharply-defined
internal border of the mesencephalic fossjx, and forms therewith
a prominent angular boss of bone, especially well marked in
Geococcyx. The'bony falx is not very strongly developed.

The ' oculo-motor (iii.) and abducens (vi.) nerves leave by a
common aperture, foi'ming a deep gi'oove across the posterior
angles of the optic platform.

The cerebral fossai lie quite in front of the cerebellar fossa, and
are more flattened dorso-ventrally in the Cuculi than in the
Musophagi.

The olfactory fossfe are relatively feebly developed.

The Fremaxilla.

The external, as distinguished from the palatal, portion of the
upper jaw is made up mainly by the premaxilla. This region of
the jaw, in the Cuculi and Musophagi, varies considerably in form.
iSTormally, it may be described as about as long as the cranium.
In Croiophaga it is slightly longer than this, in Geococcyx and
Sanrothera very much so. Tj^^ically, it may be described as
hemici'escantic ni outline, depressed from above downwai-ds, and
tapering from the base forwards. In Geococcyx and Sanrothera
it is long and pointed; in Croiophaga surmounted by a sharp,
high keel ; in the Musophagi more or less inflated.

The premaxillary poi-tion of the jaw is slightly decurved at the
tip, and the palatal surface is level with the culmen. The nasal
processes of the premaxilla fuse completely with the nasals.

A nasal hinge occurs in Turacas among the jMusophagi, but is
wanting in the Cuculi.

The Maxillo-jugal Arch.

The maxilla is indistinguishably fused with the premaxilla.
The maxillo-palatine processes are expanded horizontally, and
meeting the ventral border of the nasal septum in the middle line,
foi'm an indirectly desmognathous palate.

266 MR. w. p. PYCRAFT ON THE [Mar. 17,

Among the Cuculi, the least sjDecialisecl palate is probably to be
found in Coua and Etidynamys. In Goua the maxillo-palatine pro-
cesses are widely separated and spongy in character ; the consequent
palatal vacuity is filled by the nasal septum, which is also spongy
in character, and, fusing with the maxillo-palatine on either side,
forms an indirectly desmognathous palate. Eudynamys closely
resembles Coua in this respect, but the bony tissue is less spongy,
and the fusion between the palatine processes and nasal septum is
more complete. Cioculus, Gfibira, and Gentropus resemble Etidy-
namys. Further specialisation of the palate is seen in the still
further obliteration of the boundaries between the maxillo-palatines
and the nasal septum, and the tendency to shorten and broaden
the palatines. Forms like Geococcyx and Rhinococcyx show how
these modifications have come about, whilst in Rhamfhococcyx we
may see the maximum develoi^ment of these peculiarities.

In the Musophagi the maxillo-palatines never appear to com-
pletely coalesce in the middle line, nor does the nasal septum
descend to the level of the ventral surface. The palate is never-
theless desmognathous, since this septum fuses with the dorsal
surfaces of the maxillo-j^alatines.

No separate elements can be distinguished in the quadrato-jugal
bar.

In the Cuculi the quadrato-jugal bar is almost or quite con-
tinuous with the tomium of the maxilla : in the Musophagi, on
the contrary, it I'ises considerably above the level at its distal end.
This difierence is due to the fact that in the Cuculi the floor of
the maxillo-j)alatine process lies low, being only just raised above
the palatines at its postero-external angle, whilst in the Muso-
phagi this region is raised high above the palatines.

The Vomer, Pcdatines, and Pterygoids.

The vomer is absent in the Cuculi, vestigial in the Muso-
phagi. In the latter group it is spicular in form and fused with
the palatines, which it joins by means of a pair of very short
limbs. Anteriorly it touches the nasal septum.

The palatines, in the Cuculi, vary in shape. In the less
specialised forms, such as Eudynamys, Goua, Guira, they are
moderately long, and exposed slightly outwards. Anteriorly, from
the level of the forward face of the antorbital plate to the point of
fusion with the maxillo-palatines, the shaft of each is rod-like ;
cephalad of this point the bone spreads out into a moderately
broad blade, the postero-external angles of which are rounded off,
whilst the mesial border is produced ventrad to form a more or
less pronounced keel. In Guculus there is but little difference
in the width between the anterior and posterior moieties, and the
shaft is nearly straight.

In the heavy-billed forms like Rhamiphococcyx and Rhinococcyx
the palatines are relatively shorter, and have the maxillo-palatine
extremity laminate instead of rod-like. The skull of Scythrops, I
regret to say, is not contained in the National Collection.

1903.] OSTEOLOGY OF THE CUCULIFORMES. 2(57

In all the Cuculi, the palatines meet one another in the middle
line immediately heneath the pavasphenoidal imtrum, whieh they
gi-asp laterally through their fusion with the hemii)tery<i()ids.

In the Musophagi the palatines are relatively longer than in
the Cuculi, from which they also differ in that they trcper, instead
of broaden, rapidly as they approach the pterygoids. Furthermore,
they do not meet below the parasphenoidal rostrum, but, on the
contrary, are separated one from another by the whole width of this
rostrum '(PI. XXII. fig. 2). The dorsal surface of the pterygoid
end sends up a long and deep, incurved keel, the antero-internal
angles of which fuse with the vestigial vomer.

The ■pterygoids, in the Cuculi, are moderately long and straight,
and in some, e. g. Crotop/taga, Rhhiococcyx, G'nira, Taccocoua,
have the dorsal border raised up into a high, sharp crest, the
palatine end of Avhich embi'aces the parasphenoidal rostrum, whilst
in Eudynamys, Coiuc, Geococcyx, and Fiaya, for example, this
crest is absent. The palato- pterygoid articulation is in the form
of an obliquely transverse hinge-joint, permitting only a lateral
motion. Basipterygoid facets are absent.

The hemipterygoid element of the pterygoid is conspicuous only
in Taccocoua and Geococcyx. Herein it forms a continuation of
the dorsal crest of the shaft of the pterygoid, and rests on tlie
palatine at a slight angle. Secondary fusion between the hemi-
pterygoid and the main shaft of the pterygoid Avould reproduce
exactly the conditions of the palato-pterygoid articulation which
obtain in the Bucconidce and 2[omoiida'.

In the Musophagi the pterygoids are relatively short, somewhat
twisted, rods, bearing vestiges of basipterygoid facets. They
articulate with the palatines as in the Cuculi.

The Mandibles.

The mandible in the Cuculi has a short, blunt angular process,
and a modei-ately long internal angular process.

In Eudynamys, Citcidus, and Gtoira the lumi ai'e pierced by a
long lateral vacuity, which is partly closed by a long and slender
coronoid. In Coica the coronoid tei'minates midway aci'oss this
vacuity ; Avhilst in Taccocoua iind Centropus this vacuity is quite
open, the coronoid forming its venti-al bordei'. In Fiaya, A'hino-
coccyx, and Rhampliococcyx the ramal vacuity is absent. In
Geococcyx it is partly closed anteriorly by the hinder end of the
splenial.

The mandible of Cucidas possesses one conspicuous feature in
the presence of a triangular flange of bone formed by the deflection
of the superior border of the I'amus at the point corresponding,
in the living bird, to the gape, in the skeleton to the region
imniediately behind the lachrymo-nasal fossa. The flange,
especially conspicuous in Cucidus canorus, is also fairly distinct in
Cacomantis, and is tracesible in Fiaya and Fhanvjihococcyx. The
internal angulai- piocess is well marked in all the Cuckoos, but is
especially so in Coua and Centropus.

268 MR. ^Y. P, PYCRAFT OjS^ THE [Mar. 17,

In the Musophagi the angulay' is sharply truncated posteriorly,
and the internal angular process is short and blunt.

In Schizorhis the lumal vacuity is closed posteriorly, in Turacus
it is closed completely, by the coronoid.

The Hyoid.

Unfortunately the hyoid bones have been preserved only in one
or two of the skeletons of this group in the Museum Collection.
Years ago, when these skeletons were made, according to the
prevailing custom, the hyoid bones were not regarded as of value.

In Scythrops the basihyal is slender and rod-shaped ; the 1st
' basibranchial triangular, with concave sides, the 2nd of medium
length, styliform and tapering. The ceratobranchial and epi-
branchial are subequal in length and offer no points for special
comment.

In the Musophagi the basihyal bones are reduced to mei-e
vestiges. The 1st and 2nd basibi'anchials are short and fused.
The ceratobranchials are only slightly longer than the epi-
branchials.

iii. The Skull of the Nestling.

The ISTational Collection of nestling skulls of this group is
extremely limited, containing only one skull of Cuculus canorus
and one of Geococcyx calif ornianus. A series of skulls of nestlings
and of immature individuals of various species of Cuckoos, as
well as of Plantain-eaters, would be a useful addition to the
Collection. At present the latter contains no skeletons of nestling
Plantain- eaters and few skeletons of adults.

a. Cartilage-hones.

The hasioccipital is somewhat linguiform in shape, and widens
gradually from behind forwards. It is bounded on either side by
the latei'al occipitals, in front by the basitemporal plate. Its
posterior border forms the greater portion of the occipital condyle.

The exoccipital, or lateral occipital, viewed externally, takes the
form of a broad horizontal plate bounding the hasioccipital on
either side, and extending backwards and outwards, expands into
a fan-shaped plate, the superior half of which is wedged in
between the supraoccipital, parietal, and squamosal, whilst the
ventral, or downwardly directed, moiety forms the posterior
boundary of the tjanpanic cavity. The lateral segment of the
foramen magnum is formed by the horizontal plate of the lateral
occipital, which also, by the way, forms the extreme outer angle
of the occipital condyle. The fact that the exoccipital comes into
contact with the parietal is noteworthy : a similar relationship
obtains also in Dromceus among the Palseognathas, and will
doubtless be found among seveiul other ISTeognathine forms. The
exoccipital is only just visible, from the cranial cavity being
ovei'laid by the opisthotic.

1903.1 OSTEOLOGY OF THE CUCLXIFORMES. 269

The supraocclpital is extremely short untero-posteriorly, and
is deeply cleft in the miihlle of its superior border. Its external
lateral border fuses with the lateral occipital ; Avithin the cranial
cavity it is bounded by the epi- and opisthotic. The small size
of the supraoccipital recalls the skull of the Owls, but this of
course is but a coincidence. The character will doubtless be
found to obtain in the Kingfishers, Bucconidje, and Capitonidfe,
which have many characters in common Avith the Cuckoos.

The prootic does not appear externally. Within the cranium
it occupies considerable space, forming the floor of the mesen-
cephalic fossa, as well as a considerable portion of the lateral walls
of the basin-like metencephalic fossa. It entirely excludes the
squamosal from the cranial ca\-ity.

The epiotic is only partially ossified, and in such a way that its
boundaries cannot be made out.

The opisthotic has fused completely with the prootic.

The basisphenoid is not visible externally, being underlaid l)y
the basitemporal plate. Concerning its internal boundaries,
nothing satisfactory can be gathered from the two skulls in the
National Collection, the younger being damaged, Avhilst in the
more adult skidl it has fused with the neighbouring bones.

The cdisphenoid, in the younger of the two skulls, is not yet
completely ossified. As a result, between the external ventral
angle and the squamosal there is a wide gap, _ Avhich _ extends
inwards below the inferior alisphenoid border, dividing its outer
moiety from the basisphenoid. This gap, when viewed from
without, is seen to be filled up by the prootic. Its supero-
external angle is produced outwards to form the postorbital
process.

The orbito-sphowid is not yet ossified in these skulls.

The p)resphenoid has fused Avith the basisphenoid.

The meseihmoid, in the tAvo skulls noAV under consideration,
is yet incompletely ossified, forming but a linguiform plate ; the
interorbital septum formed by the backAvard extension of the mes-
ethmoid having been represented only by cartilage. The anterior
border of the Hnguiform plate is shai-ply truncated so as not to
extend beyond the level of the free end of the parasphenoidal
I'ostrum beloAv and the anterior extremities of the frontals
above. This truncation occurs at the cranio-facial fissure, Avhich
has cut the mesethmoid into tAvo parts : the one forming the
linguiform plate just described, Avhich ultimately forms the
interorbital septum ; the other, the septvm nasi, Avhich in these
skulls is yet cartilaginous.

The cranio-facial fissure appears to be peculiar to the neo-
gnathine (Carinate) skull ; but traces thereof are apparently to be
met Avith in the Paheognatha? (Ratita?), inasmuch as, in the skulls
of nestlings of Dromceus and Rhea in the Museum Collection, the
ossification of the mesethmoid commences, as in the Neognatlue,
by the formation of a more or less linguiform plate, and this has
its superior border deeply excised, at a point exactly corresponding

270 MR. W. p. PYCBAPT ON THE [Mar. 17,

to the vei-tical face of the posterior division of the neognathine
mesethmoid, that is, exactly in fi'ont of the antoi-bital plate of the
adult. To make the fissui'e complete, the incision would have to be
continued downwai-ds so as to completely bisect the plate, whilst
the rostrum would have to terminate at the free edge of the
posterior segment of the plate. In the adult skulls the incipient
fissui'e remains, placing the right and left olfactory chambei'S, in
the di-ied skull, in communication. The foi-mation of a complete
cranio-facial fissure is correlated with, and pei-haps consequent on,
the reduction in the length of the parasphenoidal rosti'um, which
in all the Palteognathse is of great length. In a skull of a half-
grown Casuarms scdvadorii it is, however, relatively much shorter
than in an adult C. aitstrcdis, and much overhung by the mes-
ethmoid. Whether this peculiarity obtains also in the adult of
this species I am unable to say.

The olfactory cavities do not extend backwards in the Cuculi
or Musophagi, so as to lie on either side of the mesethmoid, as in
the Tubinares for example ; the antorbital plate ai-ising in the plane
of the cranio-facial fissure.

The quadrate does not differ materially from that of the adult.

The articidare is still distinct.

b. The Membrane-bones,

The 2X(,rietal is oblong in shape, its anteiior and posteiior
borders sinuously curved, its mesial boi'der sti'aight, and its
extei'nal latei'al border being slightly convex. It is bounded in
front by the parietal, behind by the supraoccipital, and laterally
by the sqviamosal.

The frontal has its hinder border sinuously cui-ved, and
throughout the greater part of its length applied to the parietal.
Its postero-external is closely applied to, and ultimately fuses
with, the supero-anterioi- angle of the squamosal and the post-
orbital process of the alisphenoid. In the supiuorbital region it
turns downwaids and inwards to form a broad overlapping plate
articulating with the alisphenoid. Fi'om the mid- orbital region
onward it becomes band-shaped, ultimately being produced into
an outwardly directed and blunt angle vmderlying the nasal.

The squamosal is roughly quadrate in form, and has the superior
anterior angle pioduced into a lingviiform process which overlaps
the parietal and alisphenoid. Its antero- ventral angle is produced
into a small squamosal pi'ominence, the undei- suiface of which
affords the articular surface foi' the squamosal head of the quadi'ate.
The postero- ventral angle is obliquely truncated, and foi'ms the
anterior segment of the rim of the tympanic cavity. The supei'o-
posteiior angle is produced into a slight point which is wedged
in between the paiietal and lateral occipital. The hinder boi'der
of the squamosal serves to cover in the recessus tymjKinicus
sujjerior, which is foi'med by the absorption of the diploe of the
lateral occipital.

1903.] OSTEOLOGY OF THE CUCULIFORMICS. 271

Internally the aquiiniosal ;i])peai-s to l)e visi1)le only in Cacaliis
canoras and (■entro^nis, where it appears as a small triangular
plate wedged in between the probtic, parietal, and alisphenoid.

The nasal, judging from the scanty material at my disposal,
difiei's slightly in form in different geneivi, inasnnich as in Cucalus
and Centrojyas it is not sufficiently large caudad to cover the
horizontal plate of the mesethmoid. In Centro])us the hinder
bordei- is produced into a shai-p angle, and is I'ounded in Cuculus
and GeoGOCcyx. The nostril is holorhinal.

The lachrymal offei'S no special featui'es for eounnent.

The prema.clUa is fully ossified only in the skull of Ceococcyx.
Its nasal processes are long and slender, and the median suture
dividing them extends far beyond the level of the anterior border
of the nasal fossa. The maxillary and palatine processes lie close
together and parallel with one another.

The maxilla in Qeococcyx is elongated and triaiigulai- in foi-m,
and sends backwai'ds from its postero-extei'ual angle a long
slender bai- to join the jugal and <jua<li'ato-jugal. The Ixxly of
the maxilla is slightly s[)ongy and has the postero-internal angle
})roduced backwards and inwards into a maxillo-palatine process.
The botly of the maxilla lodges a Ijai-ely pei-ceptible anti'um.

The qaadrato-jugal is long and slender in Qeococcyx ; the re-
lations between the jugal and maxilla cannot be made out, the
skull having become disarticulated.

The vomer is absent.

The palatine does not differ apprecial)ly from that of the adult.

The pterygoid is rod-shaped, and prodiiced f oi'wards into a sharp
triangular spine (PI. XXII. fig. 10), which articulates with a coi--
responding facet in the mesial border of the palatine. Later this
triangular process becomes segmented off fi'om the main shaft
to form the hemipterygoid, which ultimately fuses with the
palatine. At the pi'esent stage this segmentation is faintly indi-
cated by a shallow fuii-ow on the outei- siuface of the shaft. In
the adult, whei'e the fusion with the hemipterygoid is complete,
the palato-pterygoid articulation is formed by the approximation
of glenoid sui-faces sloping obliquely backwai-ds.

The dentary and splenial do not differ from those of the adult.

The cm'onoid, angidare, and supra-angidare are all as yet
distinct.

iv. The Vertebral Column.

The vertebral colunni of the Cuculifoi'mes is not marked by the
same constancy of character which is so conspicuous in the skull.
In many respects it recalls that of the Coraciiformes, but even
among the Cuculi i-elatively wide differences obtain.

All the presynsacral vertebr:>? are heteroccelous, and all the
thoi'acic vertebi';>? ai-e fi-ee.

The cervicals of the smaller Cuculi recall those of Leptosoma.

The odontoid ligament of the atlas is ossified. The axis vei'tebra
is veiy short antei'o-posteriorly, and has the neui'al ai'ches produced

Proc. Zool. Soc— 1903, Vol. I. No. XVIII. 18

272 MR. W. p. PYCRAFT ON THE [Mar. 17,

outwards and upwards into a pair of pointed hyperapophyses.
The neural arches of the vertebrae, from the third to the fifth
inclusive, are very broad, and with or without spines. From the
third to the sixth or seventh vertebi'pe large metapophyses occui-,
those of the fourth, fifth, and sixth being especially lai-ge, pro-
jecting like buttresses from the base of the anterior zygapophyses.
The metapophyses of the third vertebra send back each a slender
bar of bone to join the hyperapophysis. A similar bar of bone
occurs also from the fourth to the seventh vertebrae, but instead
of joining the hyperapophysis, fuses with the antero-latei-al angle
of the shield-shaped expansion formed by the neural arch. The
succeeding cervicals do not afford any particularly well-marked
characters in so far as this region is concerned.

The anterior cervicals of Scythrops and Gucuhos (1 to 7) diflfer
from those of the Cuckoos just described in that they are
relatively slightly longer antero-posteriorly, that the bony bar
from the metapophysis to the hyperapophysis occurs only in the
3rd vertebra, and in that the shield-like expansion of the neural
arches is very feebly developed. Short neural spines occur in the
2nd, 3rd, and 4th vertebrae.

The 2nd, 3rd, and 4th vertebra; bear hypapophyses ; from the
5th to the 10th these are replaced by catapophyses, which from
the 11th vertebra backwards to the last cervico-thoracic are
replaced again by hypapophyses. The cervico-thoracic vertebrfe
are peculiar in that they bear more or less well-developed catapo-
physes in addition to the median hypapojDhysis, The catapophyses
never coalesce to form a carotid canal.

The cervico-thoi-acic vertebrae may be two oi- three in number.
The number of the true cervicals varies, being either 11 or 12.

The thoi-acic vertebra? are free. The 1st or 1st to 3rd, as in
Eudyncmiys and Coita for example, bear hypapophyses. Pneu-
matic foramina pierce the centra of the vertebra beneath the
transverse processes.

The neui-al spines of these vertebrae gradually increase in height
from before backwards, and are more or less quadrate in form.
Coua, however, appears to form an exception to this rule, the spine
of the 1st thoracic being almost obsolete, and thus agreeing with
the cervico-thoracic vertebi-ae ; whilst the sjaines of the 2nd, 3rd,
and 4th vertebrae are, relatively, as high as in other Cuckoos.

The last two cervico-thoracic (2nd and 3rd) have apparently
only recently become separated from the thoracic series, inasmuch
as the vertebral segments of their respective ribs have undei'gone
no reduction in length. The ribs of the third pair still i-etain
their uncinate processes.

Only one thoracic vei'tebra, in the Cuculi, enters into the
synsacrum.

The synsacrum includes from 10 to 13 vertebrae. The smaller
nuraber appears to have been due to the reduction, possibly by
excalation, of the lumbar or lumbo-sacral vertebrae, or of caudal
vertebrae, as in Geococcyx.

1903.1 OSTEOLOGY OF THE CUCULIFOBMES. 273

The most complete synsacrum is that of Coua, and is made up
as follows :— 1 thoracic, 3 lumbar, 3 lumbo-sacral, 2 sacral, and
4 caudal, making 13 in all.

Rhoj^odytes seems to possess the most reduced synsacrum : tiie
number of thoracic, lumbar, and saci-al is the same as in Coua, but
the lumbo-sacrals are reduced to 1 and the caudal to 2 : making
therefore a total of 9 vertebra? as against 13 in CoiM.

Crotophaga and Geococcyx each have 11 vertebije m this region.
In the former, the reduction is from the lumbo-sacrals, which are
only represented by a single vertebra; in the latter there are
2 lumbo-sacrals, but only 3 caudals.

In other genera, the number of vertebrfe appears constantly to be
12, composed as follows :—l thoracic, 4 lumbar, 1 lumbo-sacral,
2 sacral, and 4 caudal.

In no ca^se can distinct sacral vertebrfe be made out. ilie
sjns^cvxm-ioiRhmnjyhococcyx is remarka.ble in that the single lumbo-
sacral vertebra bears a pair of exceptionally strong ventri-latera
processes. Yestiges of these appear in Scythrojis, Coica, and

Taccocoua.

In Centropus the third and fourth pairs of ventri-lateral processes
are apparently undergoing coalescence. In some skeletons this
fusion is complete; in others traces, more or less marked, of the
original buttresses still remain.

The last two caudal vertebra? combine to form a well-marked

r)lanum anale.

There are 6 or 7 post- synsacrals— free caudal vertebrae. Ut these
the 4th and 6th bear pointed intercentra, which, however, are
completely fused with their respective centra. _

The total number of vertebra? reaches its highest m Coua

with 37. . . , T ..• • 1

The vertebral column of the Musophagi is scarcely distmguisli-

able from that of the Cuculi.

The cervical vertebra? closely recall those of Cuculus. Ihey niay
be distinguished therefrom, however, apart from their greater size,
in that the 3rd and 4th both send backwards a hnv from the meta-
to the hyperapophysis, and in that the 7th and 8th send back a
similar bar fiom the same region to the middle of the neural arch.
There are only two cervico-thoracics. The vertebra coirespondmg
to the third cervico-thoracic of certain Cuckoos, e. g. Scythrops,
becomes in the Musophagi thoracic, being joined to the sternum by
a sternal rib. The thoracic vertebra? differ from those of the
Cuckoos, in that the transverse processes send l)aclvwards from their
postero-external angles a long bony spike to overlap the ti-ansverse
process of the vertebra next behind it. These connecting-rods
are either wanting or very feebly developed in the Cuckoos, e. g.

Sc)/throps. . .

The cervical vertebra? are 12 in number, the cervico-tlioracic Z,
thoracic 6, the last being fused with the synsacrum.

The synsacrum contains 13 vertebra?, and is made up as
follows: thoracic 1, lumbar 3, lumbo-sacral 3, sacral 2, caudal 4.

18*

274 MR. W. p. PyCRAFT ON THE [Mar. 17,

The 1st lumbo-sacral shows vestiges of ventri-latei'al processes.
Only the 2nd sacral bears ribs.

There are 8 free caudals, but the 7th is nearly fused with the
pygostyle. The total number of vertebrae is 41.

I must here point out that the determination of the sacral
vertebrae in the skeletons herein described is a purely arbitrary one ;
they cannot possibly be determined with certainty without ex-
amination of the sacral nerve-plexus, and this I am unable, at the
present moment, to make. Thvis, it may be that the plexus will show
that in some cases what is here described as the last lumbo-sacral
in reality is the first sacral. If this should be the case then, of
course, the vertebra herein described as the 2nd sacral would be
the 1st caudal.

V. The Kibs.

The cervical ribs, in the Cuculi, form a series of very broad pleur-
apophyseal lamellae enclosing the usual vertebrarterial canal. These
plates extend backwards rather be3^ond the middle of the centrum.
Seen from the ventral surface, the rib is free for about half its
length. Certain of the vertebrae in the middle of the cervical
chain have this lamella pierced by a large fenestra. In Guira,
Coua, Centropi(,s, the 5th, 6th, and 7th are so distinguished ; in
Crotophaga the 6th, 7th, and 8th ; in Scythrops the 7th and 8th ;
in Taccocoua, the 4th, 5th, and 6th. fjuculus, Chrysococcyx, and
probably other smaller forms, differ from the forms just described
in that the anterior cervical ribs are comparatively long and
slender, not lamellate ; whilst from the 7th vertebra backwards
the ribs are reduced to vestiges.

The cervico-thoracic ribs number two or three pairs. The third
pair always bear uncinates.

There are five pairs of thoracic ribs, though not more than four
are attached to the sternum. The fifth pair are overlapped by
the preacetabular ilium, and may be reduced to the merest vestiges,
e. g. Rhamphococcyx and Coua, or they may be complete, as in Scy-
throps and Cuculus. They vary somewhat in form. For instance,
in Coua, Centropus, and Guira the proximal end of the rib is
extremely broad, and the shaft after leaving the vertebra almost
at right angles, curves abruptly downwards. In other Cuckoos
the broadening of the shaft is not conspicuous, and the shaft
slopes gently downwards and backwards, so that the thorax is
conspicuously broader at the articu.lation with the sternal ribs
than above.

The uncinates are broad and strong, and show a tendency to
develop a sharply defined postero-inferior angle at the junction
with the shaft of the rib ; this is especially noticeable in Coua.

The sternal segments of the 5th jDair of thoracic ribs never reach
the sternum. They may persist as vestiges, even the thoracic
segment of the rib reaching the verge of disappearance, as in
Coua and Rhamphococcyx, or they may be of considerable size,

1903.] OSTEOLOGV OF THE CUCULIFORMES. 275

and extend more than halfway down the sternal segment of the
4th rib as in So/throps.

The sternal ribs of Coua reynaucU are i-elatively longer than in
other Cuckoos, and this fact appeai-s to be correlated with
degenerate powei-s of flight. As a residt of the lengthened rib-
segments and a change in the angle formed 1iy the corpus sterni
and the vei-tebral column, to be discussed presently, the form of
the thoracic cavity (lifters conspicuously from that of other Cuckoos
and recalls features characteristic of " Ratitte." These peculiar
featui-es ai'e wanting in Coiui ccerulea.

The cervical ribs of the Musophagi closely resemble those of
the Cuculi, taking the form of broad pleurapophyseixl lamelh^.
From the 5th to the 8th these lamell* are fenestrated as in the
Cuckoos, so that the lamella appears to be joined to the centrum
by a slender bar of bone, the bar extending to the level of the base
of the postzygapophysis. There are six pairs of thoracic ribs, five
of which articulate \Wth the sternum. Tlie 6th pair are long,
but have lost connection with their sternal segments, which
remain as small spicules anchylosed Avith the sternal segment of
the 5th pair. In the Cuckoos, it %vill be remembered,^ there
are only 5 pairs of thoracic ribs, the 5th pair of which, like the
6th in the Musophagi, fails to reach the sternum.

vi. The Sterxum and Shoulder-girdle.

The sternum of the Cuculiformes recalls that of tlie Coraciai
rather than that of any other group.

The sternum of the Cuculi may he readily distinguisheil from
that of the Musophagi in that it is relativel}^ conspicuously shorter,
being indeed in many cases as broad as long. Moreover it
appears to be undergoing yet fui'ther reduction.

The form of the posterior margin of the corpus sterni and the
relative development of the sjnna externa present considerable
valuations.

With regard to the posterior margin of the sternum, this appeai-s
to be entire in Cacomantis only. Two extremely thin oval areas
indicate, however, the presence originally of a pair of posterior
lateral processes, the space between which and the median border
of the metasternum has been filled up.

In Scythrops, Cmidus, Coccystes, G^uira, Oentropus, and Croto-^
■phaga the posterior border is interrupted by a single pair of
notches, the outside of the notch being bounded by the posterior
lateral process.

All the other genera appear to have a doubly notched sternum,
but the character of the notches varies consideral^ly. In Playa,
for example, the posterior lateral process arises directly beliind the
articulation of the 4th rib, and extends outwards and backwards
to terminate in a spatulate extremity some distance cephalad of a
line passing across the free border of the metasternum. Between
this process and the metasternum is a i\\A\\\^\t processus inter medius.

276 ME. W. p. PYCRAFT ON THE [Mar, 17,

which, like the external process, fails to reach the level of the meta-
sternal border. In Rhinococcyx, Rhamphococcyx, and Taccocoua
the posterior lateral process arises, as in Piaya, immediately behind
the articulation for the 4th rib, but it forms a wider angle with
the long axis of the sternum, and carries with it the intermediate
process, so that it looks like a branch thereof. In Coita and Geo-
Goccyx the posterior lateral process leaves the sternal plate further
back than in the forms just described. In other respects these
two sterna are quite distinct.

In Coua ccerulea the posteidor lateral process is veiy broad, and
extends backwards to the level of the posterior border of the meta-
sternum, whilst the intermediate process is short and normally placed.
Coua reynavdi, with an obviously degenerate sternum, differs from
C. ccEruha in that the posterior lateral process is slender, and the
intermediate process shows a tendency to fuse with the base of
the posterior lateral. In Geococcyx the posterior lateral process
is long and slender, but fails to reach the level of the free border
of the metasternum ; whilst the intermediate process lies nearer the
posterior lateral than the metasternum. Further, the sternum is
peculiar in the great depth of the fissure on either side of the
metasternum, giving this a long and narrow shape.

A conspicuous pointed spina externa is found in Gtdra, Citculus,
Geococcyx, and Scythrops. It is vestigial in Centropiis. A small
spina interna occurs in Taccocoua.

In Rhamphococcyx the spina externa and interna are both
present, and fused to form a moderately large spina communis. In
Piaya, Rhinococcyx, and Coua there is a vestigial sp)ina communis.
In Gacomantis the two spines remain distinct, but are extremely
reduced.

Both spina externa and interna are wanting in Crotophaga.

The carina sterni is relatively largest in Cuculus (text-fig. 43,
p. 277), Cacotnantis, and Scythrops. Its free ventral border is
strongly arched, and its anterior border is deeply concave. The
antero-ventral angle of the keel afifords a firm ai^ticulation for the
furcula. In Coua, Rha^inpliococcyx, Crotopliaga, and Piaya, the
anterior border of the keel is very deeply emarginate, causing the
free border of the keel to project forward in the form of a long
style. With the dorsal surface of the style the clavicle articulates
by means of a long hypocleideum. Apparently the length of the
hypocleideum is directly correlated with the i-etreat of the carina
caudad. In the Galli this is well seen, the climax being attained
in OpisthoGomus.

The depth of the carina is extremely i-educed in Coua reynaudi
(text-fig. 44, p. 277), so much so indeed that the power of flight
in this species miist be extremely limited. In other Cuckoos the
depth of the carina is nearly or quite equal to half the breadth of
the corpus sterni ; in the species in qxtestion the greatest depth
of the keel is barely one-fourth the breadth of the sternal plate.
Correlated with the reduction in the keel and the consequent loss
of flight, is a marked change in the position of the sternum, which

1903.]

OSTEOLOGY OF THE CUCULIKORMES.

277

appeals to liave been tlii'ust forward by a considerable increase in
the length of the pcistei-ior sternal ribs. As a consequence, the
shaft of the coracoid nnd the long axis of the sternum form neai-ly
a, right angle Avith the vertebral column, and the shaft of the
scapula lies pai-allel with the vertebral axis, instead of forming an
oT)lique angle therewith. This forward movement of the sternum,
fui'thermore, has brouglit the aci-ocoracoid upwards to the level
of the neui-al crests of the vei-tebrte, whilst the distance between
the acetabulum and the fi'ee edge of the metasternum has lieen
enormously increased. In fact, the thoi-ax of tins ])ird appi-oaches
veiy closely, in the relative position of the sternum, that of the
" RatitJe.""

Text-fig. 43.

Text-fig. 44.

Text-tig. 43.— Left side view of steniiani and shouklcr-givdle of Cucnli'.s c((iioriis, to
show tile great size of the carina sterni aud the form of the posterior liorder
of the sternum.

Text-iig. ii. — Left side view of sternum and shoulder-girdle of Cona rei/naudi, to
show the degenerate carina sterni, which is actually more reduced tliau the
figure indicates. Note also the long liypocleidcum and the donhly notched
corpus sterni.

Shortening of the sternum is most marked in Piaija. Zaaclo-
stomios, and Taecocoiia.

The sternum of the Musophagi is relatively longer and moi-e
Cora ciif oi-m than that of the Cuculi, but is at the same time thoroughly
Cuculine in chai'acter. In both Titracus and Schizorhis, the only
genera represented in the National Collection, the hinder liorder
of the sternum is doxibly notched, and the notches are of consider-
able extent, the outei'most being the deepest. The posterior
lateral process is slender, and directed straight backwards. The
intermediate pi-ocess lies midway between the posterior lateral
process and the metastei-nmn. The outermost notch is about
twice the depth of the inner.

The co/yios sterni of Turacus is relatively much shorter antero-

278 MR. W. p. PYCRAFT OX THE [Mar. 17,

posteriorly than in Schizorhis. The sternal notches of Ttbracus,
it is to be noted, though preserving the same relative proportions
between themselves, are yet relatively shallower than in Schizo-
rhis with regard to their relation to the corpus sterni.

There is a well-developed spina externa, but no trace of a sjyioia
interna. In Schizorhis the spina is fiabellate and projects from a
rounded base. In Turacus it is quadrate in form. In T. huff'oni
it projects nearly as far forwards as the antero- ventral angle of
the carina, and is distinguishable from the anterior border of the
carina only through the medium of a notch. Were this notch
filled up, tire spina would disappear and the anterior border of
the keel would present a vertical face pi-ojecting far beyond the
coracoid grooves, as in certain Ooraciifoi-m birds, e. g. Cyanops.

Thecoracoid grooves overlap one another. The dorsal lip is
extensive. A median notch occupies the place of the spina
interna, and this is bounded on either side by a conspicuous
oblong glenoid surface, which articulates with a special facet on
the coracoid. In the Cuckoos the dorsal lip does not overhang
the ventral, the coracoid grooves do not overlap nor in some
cases even reach the middle line, and the oblong articular facet
on the dorsal lip foi- articulation of the coi'acoid is only slightly
developed. In the Musophagi the dorsal surface of the base of
the spina extei-na affords ai'ticulation for the right coracoid ; in
the Cuckoos this is nevei- the case.

In the Musophagi the ai-ticular surfaces for the ribs ai-e fairly
widely spaced, less so in Turacus ; in the Cuckoos these surfaces
are crowded together.

The coracoid in the Cuculi is relatively long and slendei-, being
nearly or quite as long as the sterniim. From the ventral aspect,
one of the most conspicuous features is the large procoracoid.
This forms a large oblong shelf pi'ojecting inwards and downwards
from the shaft, at about the level of a line drawn across the shaft
from behind the glenoid cavity. The acrocoracoid is lai-ge, and
not twisted on the shaft so as to conceal the foramen interosseum.
The processus lateralis is well developed and t\\e foramen sup>ra-
coracoideum is absent.

The processus lateralis hasalis may be either bi'oad and quad-
rate, with its antero- external angle produced forwards into a
spine, as in Cucuhis, Rhopodytes, Eudynaonys, Taccocoua, Rham-
phococcyx, Scythrojis, and Guira ; or narrow and directed outM'ards
and backwards, as in Gentropus, Goua, and Bromococcyx. In
Taccocoua, and to a less marked extent in Gentropus, the impres-
sion for the sterno-coracoideus muscle, on the dorsal aspect of
the coracoid, is bounded in fi'ont by a sharp vertically directed
spine.

The scapula is long and narrow, and differs somewhat in shape
in the different genera ; but the variations are unimportant, and
not sufficiently^ large to justify desciiption here. The acromion is
short and stout, save in Goua reynaudi, wherein it is reduced in
width to form a somewhat cylindrical process.

1903.] OSTKOLOGV OF THE CUCTLIFORMES. 279

The fitrcula lias long, round, slender limbs, gently arched.
There is a moderately long and styliform hy})ocleideum in Geo-
coccyx, Cotia, Rhamiihococcyx^ and Playa, for example. In Croto-
2iha(ja it is spatulate. In Centropus, Cuculus, Eudynamys, and
Caconiantis it appeal's to ])e of a degenei-ate spatulate type ;
whilst in Scythrops the h}-pocleideum appears to be wanting.

The natui'e of the combination of the elements forming the
inner wall of t\\Q foramen trlosseuvi — the acromion of the scapula,
the jirocoracoid, and the free end of the clavicle— may prove,
when exh:iustively worked out, to have some slight value from a
systematic standpoint. I propose, howevei-, here to offer only a.
few I'emai'ks, selecting a few genei'a as examples.

In Cuculus, tScythrops, Co^ia, the whole free end of the furcula
passes up cephalad of theaci'omion of the scapula and separates the
latter fi-om the pi-ocoracoid. In Guira. Geococcyx, Rharaphococcyx,
RJdnococcyx, Rhopodytes, the acromion tuiiis foi'wards so as nesirly
to join the procoracoid, and forms an opposing surface to the free
end of the clavicle, which accordingly tui-ns foi-wards to tei-minate
in a point wedged in between the acromion and pi'ocoi-acoid. Cro-
tophaga resembles Guira in this i-espect, but the fi-ee end of the
clavicle is much broader and the articulation with the procoi-acoid
moi-e developed. In Centropus the aci-omion and acrocoracoid
meet and embrace, as in a wedge, the styliform free end of the
slender fui-cula. The pi'ocoi'acoid, which is lai'ge, does not come
in contact with the scajnila.

No two of the genei^a,, however, exactly agxee, but the diffei-ences
between them ai'e too slight to be desciibed in words.

The coi-acoid of the Musophagi, though closely resembling that
of the Cuculi, is yet i-eadily distinguishable thei-efrom by the fact
that the pi-ocoracoid tui-ns downwards and outwai'ds to fuse with
the acrocoracoid, thus encii-cling the foramen ti-iosseum with a
continuous bar of bone. The j^i'ocoracoid is large and passes
insensibly backwards into the coracoid shaft, and thei-e is a
supracoracoid foi-amen. The p'oces.s«.s lateralis is well developed,
and directed outwards and ujiwai-ds into a point. In the i-egion
whei-e the pro- and aci-ocoi'acoid fuse thei-e is an elongated facet
foi- the ai'ticulation of the fui-cula. The procoi-acoid, however,
appears to take the gi'eater pai-t of the shai'e in furnishing this
sui-face. On the doi'sal or intei'nal aspect of the coracoid there is
developed a special articular surface for the dorsal lip of the
coracoid gi-oove of the sternum.

The furcula has relatively shorter, broader, and more laterally
compi-essed limbs than in the Cuculi. A hypocleideum is wanting.
The dorsal extremity bears a special projecting facet for articu-
lation with the procoracoid.

The scapula I'esembles that of the Cuculi, but is relatively
slightly broader and has a large acromial process.

The pro- and acrocoracoid meet and fuse one with the other, thus
excluding the fuicula from pai-ticipatiou in the formation of the
foramen triosseum.

280

MR, W. P. PYCRAFT ON THE

[Mar. 17,

vii. The Pelvic Girdle.

The pelvic girdle of the Cnculiformes, though recalling in
certain f'eatui'es that of the Coraciiformes, is yet quite distinct
therefrom. Within the grovip it presents a comparatively wide
range of form.

Among the Ouculi, the least specialised pelves are those of the
smaller forms, such as Cacomantis (text-fig. 45) and Chrysococcyx.

In Cacomantis the preacetabular i-egion of the ilia is sepa-
i-ated by a low swollen ridge formed by the centra of the lumbar
vertebrae. The dorsal border is nearly straight; the ventral,
external, border is deeply emarginate ; the anterior border trun-
cated and rounded, curving outwards to form a hook -shaped
antero- ventral angle. The postacetabular region of the ilium has

CD \ SO

/

Dorsal aspect of the pelvis of Cacomantis meruUniis, X 5, sliowing the widely
separated preacetabular ilia, and the broad dorsal plane of the postacetabiilar
region.

a broad dorsal plane which abuts against the long and slender
transverse processes of the vertebrae. These last, by the way, are
sepai'ated by a row of intertransverse sacral foramina, one on
either side of the column. The dorsal plane of the postacetabular
ilium is not produced outwards so as to overhang the ilio- ischiadic
fissure, and its postero- external border is markedly depressed,
not forming an upturned crest as in forms to be presently
desci'ibed. That portion of the ischium which, by its upturned
growth, converts the obturator fissure into a foramen is very
narrow ; below and behind the foramen the ischium is produced

1903. J OSTEOLOGV OF THE CUCULIFORMES. 281

backwards into an obliquely sloping bar. The 2^idn.-i is long,
slender, and much bowed and closely bound to the posterior half
of the ventral border of the ischium, lieyond which it extends for
some considerable distance. The obturator foramen is Ijounded
posteriorly by a broad bar of bone descending fx'om the ischium
and fusing with the pubis. The pectineal process is re<luced to a
vestigial condition.

In the form of the pelvis, Cuculus, Coccijstes, Scijthrops, differ
V)ut little, save in size, from Cacomantis merulinus. In Scijthrops,
however, the obturator foramen is not cut off posteriorly by the
descent of a bony plate from the ischium. In all save Cacomantis
and Cuculus the intertransvei'se saci'al foramina are more or less
completely filled up.

In Mopodytes and Emlynamys the mesial border of the extreme
anterior end of the preacetabular ilium rises to the level of the
neural ridge as in Geococcyx (text-tig. 46, p. 282). In Taccocoua this
elevated region extends somewhat fui-ther loackwards and forms
a quadrate plate, the free border of which is lightly applied to the
nem-al crest. In Crotophaga, Bhamphococcyx, Coua, and Piaya
actual fusion takes place between this plate and the synsacrum. In
Centropus and Dromococcyx the relationship between the fused
region of the ilium and the synsacrum has become still further
modified, so that the extremity of the ilium has acquired i\ T-shaped
form, the dorsal limb of the cross-piece bridging a widely open
canalis ilio-lambalis. A similar canal is of course formed in the
other cases where the ilia meet the neural ci-est of the synsaci'um,
but it is inconspicuous.

Further modifications in the form of the pelvis are the
enormous lateral expansion of the dorsal plane and the short-
ening of the pubis. In forms like G^-otophaga, Coua, or Ehain-
phococcyx, the early stages of the first-mentioned modification
may be studied. Passmg through forms like Centropus and
Taccocoua, we reach a climax in Geococcyx (text-fig. 46, p. 282).
Herein the mdth across the dorsal plane equals the length of the
synsacrum. This outward extension of the postacetab\ilar ilium
has resulted in the foi-mation of a huge ledge passing far beyond
the level of the antitrochanter, and finally, at its free edge,
turning abruptly downwards and then suddenly upwards, back-
wards, and inwards, like a pair of wings, so that the surface of the
dorsal plane acquires a peculiar saddle-shaped appearance. Seen
from the side (text-fig. 47, p. 282) or from below (text-fig. 48,
p. 282), this overhanging ledge forms an enormous penthouse
above and slightly beliind the ilio-ischiadic foramen. A similar
modification obtains elsewhere only among certain Rails. It
api^ears to be correlated with a terrestrial mode of life, these forms
dying but little.

' The shortening of the pubis is most marked in Geococcyx,
Piaya, Centropus, and Rhinococcyx, and to a less extent in Coua
and Crotophaga.

A large pectineal process occurs in Geococcyx, Coua, Centropus,

282

MR. W. P. PYCRAFT ON THE

[Mar. 17,

Rhmnphococcyx, Rhinococcyx, and Piaya ; it is small in Rhopodytes
and TacGOcoua, and vestigial or wanting in other genera.

The fovea lumbalis is very small, and the fovea ischiadica and
jnidendalis are confluent. There is a weil-mai'ked iliac recess.

Text-fig. 46.

Text-fig. 48.

Text-fi2-. 47.

Text-fig. 46. — Dorsal aspect of the pelvis of Geococcyx mexicanws, showing the
narrow preacetabular ilia, extremely broad postacetahular dorsal plane, and
large pectineal process.

Text-fig. 47. — Side view of same, to show the broad ledge-like plate of the post-
acetabular ilium, the short pubis, and large pectineal process.

Text-fig. 48. — Ventral aspect of same, showing the great overhanging ledge of the
dorsal plane. All figures of natural size.

The pelvis of the Musophagi resembles that of the more
specialised Ouculi. Though relatively vi^ide, its breadth is due
rather to the great length of the synsacral transverse processes
than to the dorsal plane of the postacetabular ilium. The pre-
acetabular region of the ilium differs conspicuously from that of
the Cuculi, inasmuch as in this group the doi'sal aspect is so much
cut avv^ay as to lie far below the level of the synsaci^al crest, save
where it rises at the extreme antero-doi'sal ansrle. In the Muso-

1 903.] OSTEOLOGY OF THE CUCULIFORMES. 283

phagi, the preiliuin rises upwards and sweeps iuwai'ds in the foi'in
of a broad veitical l)lade with a gently lounded dorsal border,
which ultimately meets and fuses with the synsacral crest, termi-
nating in a truncated anterior bordei- cephalad of the synsacral
fusion. The cancdls ilio-lumhalis is wide and spacious. The
dorsal plane of the postacetabular ilium is moderately wide, and
tapers posteiioi-ly into a blunt, slightly uptui'ned point. Its free
outer boi-der is somewhat thickened in Turacas to form a broad
lip with shai'ply defined ujjper and lower edges ; the infei'ior of
these edges, immediately behind the ilio-ischiadic foramen, is pro-
duced into a blunt spine. In Schizorhis the free bordei' of the
doi'sjil plane is shai'ply defined, so that the blunt point caudad of
the ilio-ischiadic foramen is apparent on the doi\sal aspect of the
pelvis, instead of below this level. The ilio-ischiadic foramen is
relatively much smaller in Taracus than in Schizorhis. The ischium
expands caudad into an extremely broad plate, especially so in
Titracus. The fi'ce hinder border of this plate is slightly convex.
In Turacits the obturator foi'anien is cut off" from the fissure of
that name by a descending plate of bone from the ischium. This
is not the case in Schizorhis, the foi'amen and the fissure being
confluent. The pubis is veiy long, attached foi- some distance
posteriorly to the venti-al boi-dei- of the ischium. The pectineal
process is lai'ge in both the genei-a in question.

The fovea lamhalis is larger than in the Cuculi, and the fovea
ischiadica is distinct from the fovea piidendalis, the latter being
cut up into separate compai'tments by the synsaci'al venti-i- lateral
processes. The iliac i-eceas is faiily spacious.

viii. The Pectoeal Limb.

The character of the wing is very uniform throughout the
group. The humerus only is pneumatic. Most neai-ly resem-
bling that of the Ti-ogons in its genei-al chai'acter, the limb of
the Cuculifoiiiies may be distinguished by the greatei- size of the
pectoral crest of the humei-us, which forms a linguifoi-m or even
triangular jDlate, and by the pi'ominent coUum trochlece, the
strongly bowed curve of metacarpal III., an<l the absence of a
backwardly projecting spur on the uppei- ^ of metacai'pal II.

In the humerus the coi'aco-humeral gi'oove (sidcns transversum)
is wanting. The crista superior is gently arched, rising from the
base of the titbercuhom externus and terminating about the upper
5 of the shaft. In some cases, e. g. CrotopJutya, the pectoi'al crest
may be moi-e or less triangular instead of i-ounded. There is a
very small ectepicond^dar tubercle. The pneumatic foramen is
small ; the indsura capitis sharply defined. The humerus of the
Musophagi may be distinguished from that of the Cuculi by the
fact that in the formei- the pi'oximal Ijorder of the crista superioi-
is long, low, and coneive, the distal boi-dei- shoit and sti-ongly
arched.

In the Cuculi there is considerable variation in the relative

284 ME. W, p. PYCRAFT ON THE [Mar. 17,

lengths of the segments of the wing. Thus, in Scythrops, the
brachiuni and antebrachium are of equal length, whilst the
manus is only a very little less in length than either of these
segments. The brachium is longer than the antebrachium, and
the latter is in turn longer than the hand in Coua, Rhamjohococcyx,
RhinocoGoyx, Eiidynamys, Guira, and Groto'phaga. The relative
disproportion between the length of the brachium and ante-
bi'achiuni is especially marked in Dromococcyx and Piaya. The
brachium is longer than the antebrachium, but the latter and the
manus ai'e equal in length in Centropus and Taccocoiia. In Cuculus
and Cacomantis the brachium and manus are subequal and
shorter than the antebi-achium. In Coccystes the antebrachium
and manus are subequal and longer than the brachium.

The variation in the length of the wing among the Ouculi
becomes still more forcibly illustrated when compared with the
hind limb, the difFei'ence apparently varying with the power of
flight.

In Cuculus, Coccyzus, and Cacomantis, for example, the wing
is considerably longer than the leg. In Chrysococcyx the two
limbs are subequal. In Geococcyx, Crotopliaga, and Coua the
wing is much shorter. Taking the combined length of the femur
and tibio-tarsus of each species as a standard of measurement, it
will be found that in Crotophaga the wing exceeds this leng-th
only by the distance from the middle of Ph. I. D. II. to the tip
of the wing. In Geococcyx the wing is slightly less than the
length of the standard. Coua has the shortest wing of all, it
being about \ shorter than the combined length of femur and
tibio-tarsus.

In the Musophagi the brachium is somewhat longer than the
antebrachium, but the latter is considerably longer than the
manus. The ulna, as in the Ouculi, bears prominent bosses for
the attachment of remiges.

As in the Ouculi, the collum trochlece is very conspicuous, giving
the distal end of the humei'us an obliquely truncated form. The
coraco-humeral groove is wanting. The proximal border of the
crista superiot- is veiy long and concave, the distal border short
and convex. The ectepicondylar tubercle is very small.

The Ph. of D. III. in the Ouculi sends outwards from its post-
axial border a prominent spur, this is wanting in the Musophagi.

ix. The Pelvic Limb,

The pelvic limb in the Ouculiformes has, in common with the
Psittaci and many Ooraciiformes, a zygodactyle pes. The hypo-
tarsus is complex, and the tibio-tarsus has an ossified extensor
bridge. In all save Geococcyx the limb is non-pneumatic, and in
this genus only the femur is pneumatic.

The pelvic limb of the Ouculiformes has the 2nd and 3rd
trochleas nearly equal in length, but the latter is conspicuously
larger in size. The third trochlea is placed comparatively high

1903.] OSTEOLOGY OF THE CUCULIB'CRMES. 285

up the shaft, and has the glenoid articulai' suiface t'onspicuousl>'
laterally compi'essed and twisted, so as to cross the long axis of
the shaft transversely. This limb may be readily distinguisheil
from that of the Psittaci, inasmuch as in the latter the trochle;i
of D. II. stands out at right angles to the tarso-metatarsal shaft,
whilst the tiochlea foi- D. IV. has lotated so that its articulai-
sui-face is tui-ned to look diiecth' back^\"a^•ds. Thus, the outei-
condyle comes to lie next the shaft, and, furthermore, is pi'oduced
into a hook-shaped pi'ocess of considerable size.

Amongst the Coraciiformes, probably the pelvic limb of Leyto-
soma most nearly I'esembles that of the Cuculifoi-uies, liut the
conspicuously highei- position on the shaft of the tiochlea of D. IV.,
which obtains in this last gioup, is sufficient to prevent confusion.

The til)ular ridge is confined to the uppei' end of the shaft, and
increases in depth from above downwards.

The length of the fibula ma}' vaiy considerably even in diflerent
species of the same genus. Usually it is greatly reduced, but in
Centropus toulox(,, for example, it extends considei'ably be}'ond the
middle of the tibial shaft, whilst in C. madayascariensis it falls
considei'ably short of this.

In Crotophaga, Eudyvamys, and Rhopodytes it barely extends
beyond the level of the tibial fibular ridge.

The cnemial crests are, as a I'ule, feebly developed, indeed only in
GeocoGcyx do they attain to any respectalile size. In this species
the entocnemial crest is more oi' less quadmte in foi'm ; its outer
border is convex, its infeiior concave. The ectocnemial ci'est is
triangular in form, and tei'minates in a small pointed process
immediately in front of the head of the fibula.

The shaft of the tibia is long, slender, and cylindiical, and bowed
slightly forwards. In Geococcyx the posterioi- latei'al bordei-s of
the intei-nal mesotai-sal condyle ai-e pi-oduced backwai'ds and
upwards to form a i-athei- prominent spui-.

The popliteal fossa of the femur is obsolete.

The pelvic limb of the Musophagi is less specialised in some
respects than that of the Cuculi. This is especially marked in the
character of the tarso-metatarsus. As in the Cuckoos, the hypo-
tarsus is complex, and the tarso-metatarsal shaft grooved in front ;
but the arrangement of the distal trochlea? is of a less specialised
character, and this fact is especially mai-ked in SchizorJds, the foot
of which in this I'espect closely resembles that of Leptosoma. The
foot of Schizorhis, however, may readil)- l)e distinguished from
that of Leptosoma inasmuch as in the lattei' the trochlea foi' I). IV.
bears a very sti'ongly-developed outei- lip, \\hich is pi'oduced
inwards towards Mc. I. In ,Schizorhis this trochlea is almost
indistinguishable from that of an ordinary eleutherodactyle foot.
In Turacus, however, the form of the outer tiochlea nearly
resembles that of the Cuculi, but is less mai-kedly compiessed,
and looks backwaids and slightl}- inwards, instead of being twisted
so as to cross the shaft transvei-sely. Moreovei', it is not raised
high up on the shaft as in the Cuckoos.

286 MR. w. p. PYCRAFT ON THE [Mar. 17,

X. Summary.

The isolated position of the Cuculiformes among the Coracio-
morpha? is as evident fi'om a stvidy of the osteology of the group
as fi-om other points of view. Their nearest allies, judged from
an osteological standpoint, would appear to be the Coraciidse
{Coraciince and Leptosomatince) and Bucconidfe on the one hand,
and — moi-e remotely — the Opisthocomi on the other. Their
relationship to the Psittaci, which is generally agreed upon,
would, I think, never be suspected from a compaiison of the
skeletal fi'amewoi-k. This fact is pi'obably to be explained by the
great amount of specialisation which the Pari'ots have undergone.
Concerning the association of the Cuculiformes with the
Bucconidee, a few words of explanation are necessary. Although
this connection has several times been made by the older orni-
thologists, Flii-bringer appears to be the only modern systematist
who, on anatomical grounds, takes a similar view. Gadow (3)
regards the Bucconidfe as close allies of the Galbulidte, placing the
two in the same family — Galbulidee. Beddard, on the other hand,
widely separates these two forms, and I think rightly so. He
places the Bucconidfe between the Pici and the Bhamphastidse,
but remarks that this is a family which " is at present little
known, and whose affinities are therefore doubtful. It is only
provisionally that I place them in the present position."

The claims to relationship, of the group now undei- consideration,
to the Opisthocomi are, on osteological grounds, not at first sight
very strong. A careful study, howevei', of the skulls of Cuculi
and Musophagi, and a comparison with Ojnsthocomus, will show
points of agreement which suggest affinity i-ather than convei-gence.
The Musophagi most nearly approach Oinsthocomus. The pelvis
and sternum of Opisthocomus are Cuculine,

That the Cuculidfe and Musophagidfe are very near allies there
can be no doubt. Indeed, the relationship between these two is
the only relationship about which we can speak with any real
certainty.

According to the British Museum Catalogue of Birds, vol. xix.,
the Cuculidse (suboi-der Cuculi) embraces 47 genera and includes
202 species grouped under 6 subfamilies. This scheme is
the work of Capt. Shelley. Beddard (1) recognises only 3 sub-
families, a,nd these, I think, will be found to meet all demands.
Further, it is quite open to question whether even these should
not be regarded as groups a, h, c, rather than subfamilies.

Osteologically, the Cuckoos vary far moi-e widely in respect of
the pelvis and sternum than in any other character. The great
featvire in the skixll is the uniform plan of the palate. Trusting
to this alone, one might divide the Cuckoos into two groups —
Co'ua in the one, and all the rest of the Cuckoos in the other :
yet the difference in the two palates is tiivial. The only other
characters offered the systematist to choose from ai-e the form
of the lachrymal and temporal fosste and the shape of the beak,

1903.] OSTEOLOGY OF THE CUCULIFORMES. 287

but the latter at any rate is the direct outcome of adaptation.
Probably the remarkably specialised condition of the pelvis also
is adaptive.

In the accompanying table it will be found th;it I have arranged
the Cuckoos on a plan formeil of a compromise between that of
Beddard, Imsed on the pterylography and voice-muscles, and one
based on the osteolog}' and voice- muscles. The two will be found
to disao-ree in many points ; and it is on this account that
I have published the results of my comparison, since it will, I
hope, stimulate further research by dii'ecting i-enewed attention
to the anatomy of the discordant forms.

A. Sternum with sinsflo pair of notches ; pectineal process of pelvis vestigial.

Mu-icle-foniiuiii AXV-f-, except JEudynamys which has the formula
ABXY-f.

a. Syrinx tracheo-bronchial. Coccvstes

Muscle-formula AXY+, except J Cocr'iam'
E.d,n,a>,,„s which has ABX\ + . "1 js^J'l^^^;

01^1 ^^ '^^•l^l- IScyihrops.

h. Syrinx bronchial. /- rt„„*„„7,^„„

Muscle-formula ABXr+ . \ ^^'f"^"'

New World. ^ ^"'"^-

B. Sternum with single pair of notches ; pectineal process of pelvis large.

c. Svrinx bronchial. r n .
Muscle-formula ABXY+. \ Centropns.
Old World. iBromococcyx.

C. Sternum with two pairs of notches ; pectineal process large.

d. Syrinx bronchial.
Muscle-formula ABXY-I-.

New World Geococcyx.

Old World Coua.

(^ Rhinococcyx.
„ . . , , . , I Ehampliococcyx.

e. Syrinx tvacheo-bronchial. Urococci/v
Muscle-formula ABXY-H, except j Ejiopodl/te's.

^, .K'fr^'^'\ ."'^"^" '^'^' "^^^ "*■• Zanclo'stomus.

Old \\ orld. I Taccocoua.

{Fiaya. (New World.)

D. Sternum with two pairs of notches ; pectineal process vestigial.

f. Svrinx tracheo-bronchial. , ^,
Muscle-formula AXY-f. \ CJirysococ^yx.
Old World. ^ Cacomanhs.

Mr. Beddard has sho^vn, in tabular form, the points wherein
the skulls of the Cuculi and Musophagi differ one from another.
These ai'e not many, aiad I cannot but think that they are really
of slight importance. The most striking are the differences
in the size of tlie antorbital plate, the position of the maxillary
end of the quadrato-jugal bar relatively to the tomium, and the
I'elation of the mesial l)oi'ders of the proximal ends of the palatines.
The desmognathism of the palate seems to me to be precisely
similar in both foiins. If the skull of Turacus, for example, be
compared with that of Coua, it will be found that the chief
difference lies in the moi'e spongy nasal septum of the latter,

Proc. Zool. Soc— 1903, Yol. I. No. XIX. 19

288 . MR. w. p. PYCRAFT ON THE [Mar. 17,

which descends so as to lie between and force apart the maxillo-
palatine, causing the palate to be indirectly desmognathous. In
Turacus^ the nasal septum fuses with the dorsal borders of the
maxillo-palatine, and does not descend between them, but in this
case also the palate is indirectly desmognathous. In both Coua
and Tt(,racus, but for the septum, the palate would be schizo-
gnathous. But the maxillo-palatine region in ScMzorhis, for
example, is practically identical with that of normal Cuculi.
Coua has the least specialised palate among the Cuculi ; Turacus
among the Musophagi. In the more specialised skull of both
groups indirect desmognathism appears to have taken place by the
descent of the ventral border of the septum nasi between the
maxillo-palatine processes ; resulting in the formation of a flat
palatal surface, or one with a slight median gi-oove.

The very striking differences in the form of the pelvic girdle
stand in strong contrast with the uniformity found in the skull.
How considerable these differences in the girdle are may be seen
in the figures on pp. 277, 280, and 282.

Finally, I may remark that a great deal yet remains to be done
befoi'e our knowledge of the Osteology of this group is complete.
Of the nestling skeleton of the Musophagi we know nothing, and
only a very little concerning the early stages of the Cuculi.

Only about one-third of the total numbei- of genera of Cuculi
are represented in the Museum Collection of skeletons. These,
howevei', fortunately, include all the moi'e important forms.

xi. Key to the Osteology of the Cuculiformes.
A. Skull.

Palate desmognathous, and without conspicuous anterior palatal vacuitj- ; nostrils
holorhinal, imperforate ; basipterygoid processes wanting ; vomer vestigial or
wanting ; pterygoid free ; lachrymal free ; antorbital plate large ; the postorbital
process never extending down to the quadrato-jugal bar.

A. Mesial border of palatines not meeting in middle line, leaving parasphenoidal

rostrum exposed ; with a large os uncinatum Musophagi.

B. Mesial border of palatines meeting in middle line below and concealing para-

sphenoidal rostrum Cucuxi.

Key to the Genera of Cuculidfe.

a. Beak long and pointed.

a'. Lachrymal with massive descending process ; nostrils linear near the middle
of the tomium ; temporal fossae confined to lateral aspect of skull.

Geococcyx .
b'. Lachrymal very small ; temporal fossae meeting one another to form a broad

sagittal crest SaurotJiera.

h. Beak with a high median keel ; lachrymal large Crotophaga.

c. Beak swollen, with an arched culmen.

c'. Palatines fused in median line posteriorly ' Scythrops.

d'. Palatines free posteriorly.

a". Nostrils linear immediately above tomium ; postorbital process of lachry-
mal large Rliamplwcoccyx.

b". Nostrils oval immediately above base of tomium, and bounded above or
confluent with a conspicuous fenestra ; postorbital process of lachrymal
small Mhinococcyx.

1903.]

OSTEOLOGY OF THE CUCULIFORMES. 289

d. Beak not exceeding the cvauiuin in length, of medium size, and with a more or
less strongly arched culinen.
e'. Descending process of lachrymal vestigial or wanting; temporal tossa
meeting to form a sagittal crest ; nostrils more or less markedly rem form,
being formed like those of Ilhinococci/x by the confluence of the narial
aperture with a superior fenestra.
c". Descending process of lachrymal in form of delicate incm-ved style.

a'". Sagittal crest reduced to a mere line Ehopodj/tes.

b"'. Sagittal crest broad Zanclustomus.

d". Descending process of lachrymal reduced to a stump.

Piai/a.
c". Descending process of lachrymal wanting; antorbital plate slightly

inflated Taccocona.

f. Descending process of lachrymal moderately strong, bent outwards and
backwards upon the supraorbital ])ortion; temporal fosss meeting to fonn
a broad sagittal crest.
f". Horizontal process of lachrymal longer than descendmg; pterygoid with

d> irsal keel ; interorbital fenestra large Gtiira.

q" . Horizontal process of lachrymal not exceeding descending process, and
having free orbital border sharply truncated; interorbital fenestra small,

g'. Descending process of lachr\nnal slender ; temporal fossa; meeting to form a

broad sagittal crest : interorliital fenestra very large ... Ce)itropus.
/i'. TemporarfossiE confined to lateral aspect of skull.

h". Lachrymal moderately large and not distinctly divided into supraorbital
and descending processes ;' antorbital plate spongy; interorbital fenestra
large ; maxillo^palatines not meeting in middle line, but palate edged by

nasal septum Coua.

i" . Lachrymal moderately large, with slender descending process; nostrils

cricular, just above base of tomium Coccystes.

J". Lachrymal vestigial ; mandible with broad flange at gape.

c'". Temporal fossai present Cttculus.

(i'". Temporal fossa; obsolete Cacomantis.

B. Vertebrae \

All the presynsacral vertebra; are free and heterocoelous ; neural arch of atlas
broad ; hypapophvses blade-shaped, not extending beyond 2nd thoracic vertebra.
Axis vertebra not conspicuously shoi'tened antero-posteriorly, and not having
conspicuously upturned hyperapophyses.
A Lateral borders of thoracic vertebraj terminating in a sharp backwardly directed

spike MusoPHAGi.

B. Lateral borders of thoracic vertebrae not produced into spikes. CuccLi.

C. Pectoral Girdle and Sterxum.

Coracoid with a well-developed procoracoid; clefts of the posterior notches of

sternum never extending as far forwards as the level of tlie articulation ot the ribs;

furcula large.

A Procoracoid fused with acrocoracoid ; furcula with a well-marked facet for
articulation with the fused pro- and acrocoracoids ; anterior border ot carina
sterni continued forwards to level of spina e.xterna M usopuagi.

B Procoracoid not fused wit)\ acrocoracoid ; furcula without articular facet for
acrocoracoid ; anterior border of carina sterna not continued forwards as tar as
free end of spina externa CocuLi.

1 The presynsacral vertebrx .A' the Cuculiformes are scarcely to be distinguished
from those of the CoraciituB and Leptosomin<e.

19*

290 ■ MR. AY, P. PYCRAPT ON THE [Mar. 17^

Key to the Genera of Cuculidaj.

a. Sternal plate with spina communis.

a'. Posterior border with two pairs of notches.
a". Hj-pocleideum styliform, free end onlj- overlapped by carina sterni.

Rlia ni'pliococcyx.
h". Hypocleideum styliform, overlapped by carina sterni. < ^ tinoeoccyx.

c". Hypocleidemn oval, overlapped by carina sterni < ■pl°^'^ '•'-' ^*"

h'. Posterior border with single pair of notches Cocetjstes.

h. Sternal plate with spina externa only.
c'. Posterior border of sternal plate with two pairs of notches.

d". Hypocleideum underlapped b}' carina sterni Zanclostomus.

e". Hypocleideum with free end onlj' underlapped by carina. Geococcyx.
d' . Posterior border of sternum with a single pair of notches.

f". Hypocleideum obsolete ; procoracoid small, affording articulation for
scapula.

a'" . Furcula large and stronglj' arched Scythrops.

V". Furcula slender and slightly arched.

a*. Posterior lateral process broad Tludynamys.

b*. Posterior lateral process narrow Ghiira.

g" . Hypocleideum obsolete ; procoracoid large, not affording articulation for
scapula Centropus.

c. Sternal plate with spina interna onlj' Taccocoua.

d. Sternal plate without spines {nTomoToccyx.

e. Sternal plate with spina externa and interna distinct.

h" . Posterior border of sternum with a single pair of notches.

C'ucuTus.
i". Posterior border of sternum with two pairs of notches . Chrysococcyx.

D. Pelvic Girdle.

k. Preacetabular ilium with stronglj^ arched border meeting in the median line abova
the synsacral spines to form a large canalis ilio-lumhalis . . . Musophagi.

B. Preacetabular ilia \^'ith dorsal border never meeting in median line above syn-
sacral spine CtrcuLi.

Key to the Genera of Ouculidfe.
a. Presynsacral neural arches broad and without neural spines ; ilia widely

separated. _ i Cacomantis.

a. With large intertransverse sacral toramma < n ,„.,i „

. No intertransverse sacral loranuna.

a". Length not exceeding 1 inch Chrysococcyx.

h". Length not less than 2h inches Scythrops.

h. Presynsacral neural arches bearing a high median ridge; no intertransverse
sacral foramina.
c'. Pectineal process vestigial.

c". Antero-dorsal angles of pre-ilium not meeting in mid-dorsal line ; pubis

veiy long JEudynamys.

d". Antero-dorsal angles of pre-ilium meeting in middle r n^„4.„^i „

line; pubis short {Guira! ^'''

d . Pectineal process large. ^

e". Iscliio-pubic fissure slit-shaped and very narrow.
a'" . Antero-dorsal angles of pre-ilium produced into ^ (Je^^frnvus

narrow bars meeting in middle line , ^ Dromococ'cyx.

V". Antero-dorsal angle of pre-ilia produced inwards to form a quadrate
plate embracing neural spines; postacetabular ilia produced into an

enormous saddle-shaped plate Geococcyx.

f". Ischio-pubic fissure extremely reduced.

c'" . Antero-dorsal angle of pre-ilia produced into a quadrate plate with con-
cave superior border just reaching neural spine ... Taccocoua.
d'". Antero-dorsal angle of ilia produced into rounded ^ -oi, ■,,„„„„„,,,,

angles, not reaching the neural spines | Blwpodytes.

e'". Ischio-pubic fissure veiy wide, closed posteriorly... Rhampliococcyx.

1903.] OSTEOLOGY OF THE CUCULIFORMES. • 291

Referexces.

(1) Beddard, F. E.— Classification of Birds. 1898.

(2) FiJRBRiXGER, jVI.— Vei-gleich. Anat. d. Bnistscluilterapparates.

Theil V. : Vogel. 1902.

(3) Gadow, H.— Bronn's Thier- Reich. B.l. vi. Vogel. Syst. Tlieil.

1893.

(4) Parker, W. K.— Tmns. Linn. Soc, 2n(l ser. Zool. vol. i., 1879.

xii. EXPLANATION OF TLATE XXII.

Fi"' 1 Ventral aspect of the stuU of Coua reynaiuli, showing the most primitive

'" ' fomi ot the indirt'ctly desmoirnathxis skull of the Cuculida'. Herein the

inuxillo-palatines do not meet in the middle line, but the palate is bridged

by the swollen nasal septum. Basipterygoid processes are wanting. Ihere

is no voiuei'. , ^ . , • ^^ • 3- J.^ ^

Fi"- " Ventral aspect of the skull of Tiiracus htiffom, showing the indirectly desmo-

"■ "" >'nathous t vpe of palate of the Musophagidic. The maxiUo-palatines do not

meet in the middle line, l)ut the palate is bridged b}' the fusion ot the yenti-al

border of the nasal septum with the dorsal aspect of the maxillo-palatmes.

The vomer is wantins, but present iu a vestigial condition in Schizo7-his.

Note the appearance'of the parasphenoid between the palatines.

Fi"- 3. Dorsal aspect of the skull of Cuua reynaudi, showing the absence ot a nasal

"' ' hin«-e and the ceiural character of the roof of the skull.
Fi- 4. Dorsal aspect of "the skull of Turacus leucotis, showing the nasal hinge and

""' " the o-eneral conformation of the roof of the skull.
Fig. 0. Dorsal aspect of the skull of SaurotJiera vetula, showing the long upper jaw

and the large temporal fossae. , ■ ^^ e e >.\

Fi" 6 Lateral aspect of the skull of SchisorJiis zonura, showing the form ot the

°' " narial apertures, the os imcinatnm and the general form of the skull.
Fi"- 7. Lateral aspect of' the skull of RJiinococcijx calorhi/nchus, to sliow the form

''' ' of the narial aperture and the ■reneral form of the cranium.
Fig. 8. Lateral aspect of the skull of a nestling Centropus, showing the arrangement
of the bones. Outer view. , ,, , • ^, *

Fi"-. 9. Lateral aspect of the inner view of the same skull, showing the separate

arrans;ement of the bones. ,. ^ x u ^i,

Fi"-. 10. The distal end of the pterygoid of the nestlmg Centropus, to show the
hemipterygoid.

^Explanation of Letters.

a?<i.=alisphenoid. ».7j.=nasal hinge.

a.r.. = antorbital plate. qp.=opisthotie.

apv =anterior palatine vacuity. o.m.=os uncinatum.

Cc.=basioccii.ital. _^j.=parietal.

6.s.=basisphriu.id. ;p.a.=palatine.

c = occipital condyle. ^aj'. = parasphenoidal rostrum.

ep.o.=epiotic. ^.o.i).=postorbital process.

e^.o.=exoccipital. ^7-o. = prootic.

/:/•. = tloccular fossa. p^.=])terygoid.

fr. = frontal. 'i- = quadrate,

.■ny =bpminterv!roid. s.e.=sagittal crest.

A.««.=hemiptervgoid. s.e.=sagittal crest.

?.=lachrvmal. s.o.=supraoccipital.

meatus intevnus. s(/.= squamosal,

mesethmoid. ^/. = temporal fossa.

me.
mes.
««. = nasal.

292 MR. H. SCHERREN ON [-^W- ^l^

April 21, 1903.

Dr. Henry Woodward, F.R.S., Vice-President.
in the Chair.

The Secretary read the following report on the additions made
to the Society's Menagerie in March 1903 : —

Tbe registered additions to the Society's Menagerie during the
month of March were 67 in number. Of these 14 were acquired
by presentation, 43 were received on deposit, and 10 in exchange.
The total number of departures during the same period, by death
and removals, was 89.

The following papers were read : —

1. Linnseus and Hunter on Feather-Tracts.
By Heney Scherren, F.Z.S.

[Received March 12, 1903.]

(Text-figure 49.)

The credit of using the feather-tracts of birds as a means of
classification belongs undoubtedly to Mtzsch, whose results, edited
after his death, by Buraieister, were published at Halle in 1840
under the title ' Pterylographie'. An English edition, tiunslated
by W. S. Dallas and edited by Dr. Sclater, was brought out by
the Ray Society in 1867. Pterylosis, or the distribution of these
feather-tracts, is, according to Prof. Newton, " of prime taxonomic
importance in Ornithology, though more in the investigation of
small than of lai'ge groups." This also seems to have been the
opinion of ISTitzsch himself, who, however, was not aware that
anything at all had been done even in noting the existence of
such tracts and of the featherless spaces which he called apteria.
In his Introduction he says : —

I maj% therefore, flatter myself with the hope of awakening the interest
of naturalists by the announcement of my new results, and, bj' the enumera-
tion and detailed description of the feathered regions of the bodies of birds
to which I give the name of feather-tracts {joterylce, Federnjlitren), of
proving that these, new and surprising as they may appear to many on the
first glance at my figures, really furnish equally significant and important
characters for the certain and natural discrimination of the families of
birds.

Professor Newton (' Diet. Birds,' Introd. p. 63, note 1) says that
the only men before Nitzsch's time who seem to have noticed
feather-tracts wei-e the gieat John Hunter and the accurate

1903.] FEATUER-TRACTS IX BIROS. 293

Macartney. The observations of tlie latter were published in 1819
(Rees's Cyclopaedia, article Feathers) : —

Although the common feathers cover the whole body, thej- do not
gi-ow from every part of the skin ; they ave thickest upon the shoulders and
loins, along the underpart of the neck and breast, and do not exist upon
the lateral lines of the neck or breast, or about the un^bilicus. This
arrangement, and their being directed downwards and backwards, allows
them to cover the body more neatly, and to remain unruffled during the
motions of the bird.

The observations of Hunter appeared first in Owen's ' Catalogue
of the College of Surgeons ' (vol. iii. pt. ii. p. 311), dated 1836 : —

Although the feathers of birds appear to be an entire and uniform
covering, they do not arise equally from every part of the body, but only
from such parts of the skin as are least liable to be affected by the motion
of the contiguous parts, such as the motion of the limbs.

Hunter, however, seems to have done something more, and to
have discriminated the feather- tracts, for, on the page quoted
above, we read : —

To these groups or thickets of feathers I shall give particular names,
taken from their situation.

It is, of course, only a coincidence that Hunter used the terms
" thickets of feathers," and that Nitzsch, in his Essay, ' Pterylo-
grnphia? Avium, Pars prior,' 1833, chose a very similar name,
j)tertjlce, which Prof. Newton has translated " feather-forests." It
may perhaps be of service to recoi-d the fact.

If the date of Hunter's observations be taken at about 1785, in
which yeai- he built his Museum, an eai^lier notice of the feather-
ti-acts exists by at least twenty years. This occurs in the treatise
By Linnieus, ' Fundamenta Ornithologica,' presented at Upsala by
A. P. Bf>?ckmann, May 4, 1765 ( Amcenitates Acad. vii.). Premising
that the feathers ai-e ai'ranged in the form of a quincunx, the
author pi'oceeds : —

Nuda vero cutis est (h. e. pennarum quinciince uon perforata ant tecta)
utrinque ad colli latei-a, a capite interscapulimn versus, et ab axillis per
latera pectoris ad iuguina usque, atque per femora postica, remotis iutegu-
mentis, videnda.

This is as cleai- as language can make it. If, however, any
doubt could exist, this would be at once removed ])y reference to
the plate (see text-fig. 49, p. 294), where, in a schematic bird, the
contour-feathers are ai-ranged in quincvuices, and there the bare
space running from the neck to the femur is connected with that on
the opposite "side by the naked patch in the interscapular region.
The naked spaces, as figured, would, if looked at from above, forn\
a quincunx ; and the text i-ecalls a passage in the ' Gai-den of
Cyrus,' where Browne writes of the quincuncial arrangement in
the " feathery plantation about liirds." Although the primary
reference Ijy Browne is to the papillse in the feather- tracts, I suggest
(and in this I am glad to have the support of Mr, Southwell)

294

MR. OLDFIELD THOMAS ON

[Apr. 21,

that the passage alluded to shows that Sir Thomas Browne was fully
aware of the distribution of the feathers in well-defined i-egions,
and that these difi'ered in position and extent in different birds.
Moreover, I believe the passage in the ' Amcenitates ' shows
that Linnteus knew of Browne's work. Besides the reference to
the quincunx already quoted, there is yet another — " Pennfe qufe

Text-fig. 49.

Topographical diagram showing feather-tracts and bare spaces in schematic bird.
(Reduced from ' Amcenitates Academics,' vii. tab. i. fig. A.)

(preeter alas et caudam) reliquum corpus servant, in quincicncem
digestsg sunt " ; and his phrase " mirando nunquam satis artificio,
quod in Colymbi prfesertim corio, alutaiiorum arte prseparato . . .
est conspicuum," is strangely reminiscent of the note of the famous
Norfolk natiu'alist : " Elegantly conspicuous on the inside of the
stripped skins of the dive-fowl."

by Capt. H. N. Dunn,
By Oldfield Thomas,

2. On some Mammals collected
R.A.M.C., in the Soudan.
F.R.S.

[Received March 11, 1903.]

The National Museum owes to Capt. H. N. Dunn a collection
of Mammals, mostly small, made by him in the Egyptian Soudan
in the course of 1902. Among these no less than five prove to
need new names, thus showing how much still remains to be done

1903.] MAMMALS FROM THE SOUDAN. 295

in studying the mammals of this interesting region ; and it is
hoped that other officers Avill follow Capt. Dunn's example.

The localities at which collections were made are mostly luther
to the southward of Khartoum, the majority of them being in
Kordofan, a province hitherto almost entirely inn'epresented in
our collections.

Besides the earlier Awitings of Sundevall ^ and Heuglin - on the
Mannnals of this region, reference may be made to a paper by
Mr. de Winton on a collection from Shendy \ to the north of
Khartoum, to the same author's work ' in conjunction with the late
Dr. J. Anderson on the Mammals of Egypt generally, and to a
small papei- ' of my own on Mr. HaAvkei-'s collection fiom the
Fashoda region of the Nile.

1. ROUSETTUS STRAMINEUS Geoff.

4. c?- Khartoum. 15 August, 1902.

2. HiPPOSIDERUS CAFFER Suud.

65. c?. El Obeid, Kordofan. 22 October, 1902.
ff. caffer had already been recorded from this region by
Temminck '\

3. Megaderma froxs Geofl'.

9. Wad Medina, Blue Mle. 21 September, 1902.

4. Taphozous perforatus Geofl".

1, 2, 3. c? $ $ . Khartoum. 28 June, 1902.

5. Canis axthus soudaxicus, subsp. n.

60. $. El Obeid. 2 October, 1902. (B.M. No. 3.2.8.8.)
Type.

119. Habessa Wells, W. Kordofan. 7 December, 1902.

The eastern i-epresentative of the Senegal C. anthus F. Cuv.
Closely allied to the typical foi-m, but palei- and with mai-kedly
smaller teeth.

General characters as in true C. anthus. Colour rather paler,
a clear sandy buft". Hairs of back and of tail bi'oadly tipped with
black. Skull, as may be gathered from the measm'ements below,
smallei' and moi'e delicate.

Dimensions of the type : —

Head and body 650 mm. ; tail 230 ; hind foot (s. u.) 137 ; ear
80.

Skull: basal length 136; zygomatic breadth 78; length of

1 " Oin Professor J. Hedenborgs insamlingar at' Dasgdjur i NordiJstra Africa,"
K. Vet.-Ak. Hamll. 1&42, p. 189.
- Keise N.O.-Afrika, ii., 1877.
3 Nov. Zool. viii. i>. 397 (1901).
•• Anderson & de Winton, ' Manim. Eijypt,' 1902.
* Ann. Mag. N. H. (7) viii. p. 273 (1901).
^ Fide Anderson, ' Mamm. Ecypt,' p. 102.

296 MR. OLDFIELD THOMAS ON [-^P^^- 21,

nasals (diagonally) 49 ; intei'oi'bital breadth 26 ; breadth across
postoi'bital pi'ocesses 37 ; breadth of biain-case 49*5 ; palate,
length 72*5 ; breadth between outer corners of p/ 44.

Teeth: length of p.^ 9'6, of -p.^ (on outer edge) 14*2, of m.^ 11,
of m.^ and m.^ combined 16*8 ; breadth of m.^ 14, of m." 10.
(Below), length p.3 87, of p., 10, of m.^ 17-3, m., 8-1.

JIab. and tt/pe as given above.

This is the Jackal figured by Cretzschmar ^ from Riippell's
specimen as C. anthus, but is cleai'ly at least subspecifically
difiei-ent from that animal. Mr. de Winton has recently shown ^
that none of the earlier names of Hemprich and Ehrenberg or
other authoi'S apply to this form, and I therefore venture to bestow
one on it.

6. YULPES VULPES ^GYPTIACA Sonn.

40. $ . Khartoum. 4 September, 1902.

7. VuLPES PALLIDA Oretzschm.

8. c? (young). Wad Medina, Blue Nile. 18 September, 1902.

8. ICTONYX PRENATA Sund.

118. d" . Gebel Haraza, W. Kordofan. 6 December, 1902.

The rediscoveiy of this species is of interest, as there has always
been some doubt whether it was or was not the same as the more
northern /. lyhica. It proves to be readily distinguishable by its
smallei' size, by certain differences in its body pattern, and,
especially, by the absence of the black ends to the caudal hairs.

9. DiPODiLLus STiGMONTX Heugl.

5. $. Khartoum. 19 August, 1902.

A topotype of the species. As ali'eady noted '\ the specimen in
the Stuttgart Museum marked Meinones stigmonyx does not agree
with Heuglin's description of this animal, and is more like that of
his M. dongolanus.

It is to be observed that the j)resent Gei-bille and the next one
are so extremely alike, that it is almost impossible to distinguish
them apart except by an examination of the soles and skulls.
There is, however, a darker median area (" Scheitel und Biicken-
mitte sattei- gefarbt ") in the Dipoclillus not present in the
Gerbillus, and this confiims my previous allocation of the name,
which in any case having once been made should be adhered to.

10. Gerbillus agag, sp. n.

96. d. Agageh Wells, W. Kordofan. 17 November, 1902.
(B.M. No. 3.2.8.11.) Ty2ye.

A small species of ti-ue hairy-footed Gerbillus, with compara-
tively shoi't tail.

1 Atl. Riipp. Reise Mamm. pi. 17 (1826).

2 Anderson & de Wiuton, Mamm. Egypt, p. 213 (1902).

3 Ann. Mag. N. H. (7) viii. p. 276 (1901).

1903.] MAMMALS FROM THE SOUDAN. 297

Size small, and feet .short. General colour above soft sandy
buff, .slightly lined on the back with the dark tips to the hairs, but
without an}' marked dai'ker dorsal area. Along the l)ack the bases
of the hairs are plumbeous, but laterally, still within the .sandy area,
the hairs are broadly ringed with white sul)terniinally, though this
colour does not show on the sui-faee. Under surface pui-e sharply
defined white as usual. Cheeks, a prominent patch aliove and
behind each eye, and another behind the ear white. Whole of
fore limb white, hind limb with a s;indy line down its outer side,
the innei- side and whole of feet snowy white ; palms and soles
thickly hairy. Tail short for this group, pale .sandy above,
darkening towards the pencilled end ; white below.

Skull unfortunately broken in the single .specimen. Molars
markedly smaller and lighter than in the common Soudanese

Dimensions of the type : —

Head and body 87 mm. ; tail 100 ; hind foot (s. u.) 24 ; ear 11.
Length of upper molar series 3"7.

Hah. and ti/pe as given above.

This little Gerbille is distinguished from its neighbour
G.2iygc('ryus by its smaller size and shoi'tei- tail. Its close resem-
blance to Dipodillus stigmonyx has ali-eady been noted.

1 1 . Arvicanthis dujtni, sp. n.

103. S . Kaga Hills, W. Kordofan (about 120 ndles W. of El
Obeid). 20 November, 1902.

" Dug out of I'eddish .sandy ciiltivation soil, from among the
natives' crops of dukhan." — H. X. D.

A many-.striped species of the A. harharus group; allied to
A. zehra, but smaller, paler, and with the light and dai'k stripes less
conti'asted.

Size small, the smallest of the group. General pale ground-
coloui' buf}', the latei'al dai'ker stiipes Ijrown instead of Ijlack.
Head coarsely gi-izzled bnfiy and brow^n. Centi-al dorsal stiipe
beginning on the cro\vn, veiy nai-row, blackish, but not so deep a
black as in ^. zebra; outside this there are on each side five
unintei'rupted bufiy stiipes, sepai'ated from each othei- ))y broad
brown bands, each of which is divided down the centi'e into two
by an intei-rupted band of light, an ai-rangement essentiall}' as in the
othei' species. The light spaces aie throughout clear buff, and the
dark lines brown, a clear butf}' line pa.ssing along below the outer-
most daik line and edging the pure white of the lielly. \\\A. zebra
the outer lines at least are white, only those near the spine being
buffy. Eye-ring buffy. Eai's dull ochraceous, without darker
marking. Arms and legs pale buffy, becoming white on the
fingers and toes. Tail well-haired ; dull ochraceous aliove, with a
naii-ow and inconspicuous mesial line of black ; whitish below.

Skull conspicuously smaller than in A. zebra and the other
species of the group, with rather largei' l)ulkv ; incisors naii'ower,
but molars I'ather largei- in propoition.

298 MR. OLDFIELD THOMAS ON [Apr, 21,

Dimensions of the type : — .

Head and body 90 mm. ; tail 70 ; hind foot (s. u.) 23 ; ear 14.

Skull — greatest length 28 ; basilar length 22 ; zygomatic breadth
13; nasals 10x3*4; interorbital breadth 4-4; breadth of brain -
case 12 ; palate, length 12*2 ; diastema 6"7 ; palatal foramina 5*6 ;
leng-th of upper molar series 4*9.

Type. Old male. B.M. ISTo. 3.2.8.15. One specimen only.

A Fashoda stiiped rat presented by Mr. K.. M. Hawker in
1901 being identified as lirvicanthis zebra, the present species
may be readily distinguished from it by its much smaller size
and more buffy coloration.

Capt. Dunn tells me that this I'at was very common in the
cultivated fields of the natives, burrowing in their crops of dukhan.
It is a vei'y handsome and distinct species, and I have much
pleasure in connecting his name with it.

12. AcoMYS WITHERS Yi de Wint.
73. S . Katul Hills. 30 October.

13. Leggada tenella, sp. n.

7. Old 2 . Roseres, Blue Mle. 14 September, 1902. (B. M.
No. 3.2.8.13.) Type.

" Found in burrows in cornfield ; had 7 young in the womb." —
H. lY. D.

A veiy small species of the L. minutoides group.

General colour of cheeks and sides a clear sandy or ochraceous
bufi", with a distinct darker median dorsal area, commencing as a
narrow well-defined line on the nose, broadening on the crown,
and extending, though less sharply defined, all down the back,
and dying away on the rump. Under surface pure sharply
defined white. No lighter raarkings round eyes. Ears small,
evenly rounded, grey, their edges faintly whiter ; a large and
prominent white patch behind and below their posterior bases.
Fore limbs wholly white ; hind limbs with a narrow line of the
body-colour continued down on the hindei' side to the heel, other-
wise white. Tail about as long as the body without the head,
brown above, inconspicuously lighter below.

Skull smaller and narrower than in the Cape L. iniinutoides,
with square and well-defined supraorbital edges. Palatal foramina
ending level with the anterior fourth of m.\ Posterior palate
elongated, its hinder edge about equidistant from the last molars
a.nd the fi'ont of the bullse.

Dimensions of the type : —

Head and body 50 mm. ; tail 35 ; hind foot 11"5 ; ear 9.

Skull: greatest length 17' 2; basilar length 14; zygomatic
breadth 9 ; nasals, length 6"3 ; interorbital breadth 3 ; brain-case,
breadth 7*5 ; palate, length 9*3 ; palatal foramina 4*0 ; diastema
5"0 ; length of upper molar series 3*0.

Hah. and tyj^e as given above.

1903.] MAMMALS FROM THE SOUDAX. 299

This beautiful little species is readily characterised by its strong
sandy colour, the marked dorsal darkening, and elongated palate.

14. Jaculus gordoni, sp. n.

85. $ . Gebel Agageh, \V. Kordofan. 12 November, 1902.

104. 105. c? 6 ■ 'K.iiga Hills, W. Kordofan. 20 November,
1902.

106. c?- Gebel urn Durragh, W. Kordofan. 25 November,
1902.

Allied to J. jaculus Linn., l)ut larger and difterently coloured,
and with lougei- eaivs.

Size rather greater and build stouter than in J. jaculus.
General colour above, as compared to the yellowish "bufi"" of
J. jaculus, darker, and nearly approaching to " vinaceous buft" "
of Ridgway. Laterally, the dar-k colour seems to pass i-ather
sooner into' the pure white of the under surface. White markuigs
more extended than in /. jaculus, the cheek, supraorbital, and
postauricular white patches all large. White hip-stripe large,
weakened in colour by a faint bufty or brownish sprinkling. Fore
limbs wholly white. Hinder aspect of thighs like back. Fine hairs
of feet silvery white, the terminal half of the long digital hairs
sandy. Tail of tlie usual pattern, its basal portion isabelline bufty
above ; black subterminal band rather over an inch in length ;
white tip :|~| in.; longest hairs at end of tail 16-17 mm. in
lengi.h.

Skull shaped quite as in J. jaculus, but larger and heavier
throughout.

Dimensions of the type, measured in the flesh : —

Head and body 120 mm. ; tail 200 ; hind foot(s.u.) 63 ; ear 25.

Skull : gi'eatest length in middle line 34 ; basilar length 27-5 ;
zygomatic breadth 24"; tympanic breadth 24-7 ; length of nasals
oil outer edge 12*5; interorbital breadth 12-8; interparietal
5-7 X 8-7; palate, length 17-3; palatal foramina 4*7; diastema
9-6 ; length of upper molar series 5*1.

Hah. (of type). Kaga Hills, W. Kordofan. Also occurring at
Onidui-niiin.

Ti/2)e. Old male. B. M. No. 3.2.8.16. Original number 104.

Tliis Jerboa, which I have named in memory of the famous
General Gordon, difters decidedly from the ordinary Egyptian
J. jaculus by its larger size and difi'erent colour. The Museum
had previously received a specimen of it from Omdurman,
collected by Mr. W. L. S. Loat, but as that was young and without
skull, it could not be desciibed.

The only other species which need be referred to is Dipus
microtis Reichenow ^ from " Samar, in Nord-ostafrica." That was
founded on a voung specimen, but the description of its teeth
shows that it was s'utticiently adult not to be the young of the
present form, than which it is very much smaller ('' Lauflange

1 Zool. Aui;. X. p. 369 (1887).

300 ON MAMMALS FROM THE SOUDAN. [Apr. 21,

35 mm."), with shorter ears, and with the remarkable character
(if not clue to accident) of having no white at the end of its tail.

15. Lepus ^thiopicus H. & E.

6. c?. Shendy. 1 September, 1902.

92.93.94.97. d 6 2 2 ■ Agageh, Kordofan. November 1902.

39. c? . Wad Medina, Bhie^Nile.

16. Procavia ruficeps H. & E,

77. c? • Kaga Hills, Kordofan. 2 November, 1902.
78.79.80.83.88. Agageh Hills. November.

17. Oryx algazel dammah Cretzschm.

(5 . Kordofan,

Practically a topotype of the subspecies.

When the ' Book of Antelopes ' was published, owing to the
predilection of the senior author for ''well-established" names,
the term leucoryx was used for the species to which Lichtenstein
erroneously transfei'red that name, while the ti'ue leucoryx of the
Persian Gulf was called by Gray's name heatrix. To put matters
more in accordance with modern ideas on nomenclature, the
Scimitar Oryx should bear the name algazel Oken ; but as that
name was founded on Cuvier's figure of a Senegal specimen, and
it is practically certain that the Eastern Soudanese foi'm is at
least subspecifically distinct from the Western, a second name is
required, and this we find in the " Antilope dammah " of Riippell
and Cretzschmar. A. dammah was erroneovisly identified by
Riippell, and following him by Sclater and myself, with the Beisa,
but by its locality (" die gi-ossen Steppen von Haraza ") is clearly
shown to be the Scimitar Oiyx. This is fortunate, as the name
dammah, being earlier than heisa, would have had to be used in
the latter's place had Riippell's identification been correct, but
now the Beisa is left with its familiar name unaltei-ed.

Briefly put, the nomenclature is as follows : —

Scimitar Oryx.

Oryx algazel Oken.

0. leucoryx of authors generally, not of Pallas.

Western form : —

Cemas algazel Oken, Lehrb. Nat. iii. pt. ii. p. 741 (1816) ;
ex L'Algazelle, F. Guv. H. N. Mamm. i. pi. 376 (1819).
(Senegal.)

Eastern form : —

Antilope dammah Cretzschm. Atl. Riipp. Reise, Mamm.
p. 22 (footnote), 1826. (Haraza, Kordofan.)

White Oryx.
Oryx leucoryx Pall.

Antilope leucoryx Pall. Spic. Zool. xii. p. 17 (1777).
Oryx heatrix Gray, P. Z. S. 1857, p. 157, and of authors
generally.

i.pr.^ 2/

^-

P.Z.S.1903,Vol.I.PLXXlir,
3 4-

, y

\ I

', Nfc^t

^'^ P '" a.TTi'.l.^^

d.&.

i .'^

'. , /. ;^

a.TTh.l

*^ cc.w.

ppr

5

Lpr.

4- •

, /^-?"0 •

.' •'. •;>• ' •. • < . .

?■

a.m.

Bale &, Danielsson L**^ lith. .

POLYCLADS PROM THE STRAITS OPMALAGGA /

1903.] ox THE POLYCLADS OF THE " SKEAT EXPEDITION." 301

18. Gazella ruficollis H. it; E,

61. (S ■ Gebel Tneis, 90 miles E. of Omrlin-mrtn. 9 October,
1902.

This fine Gazelle is a most acceptable acVlition to tlie National
Collection, in which the species is still badly i-epresented.

19. Gazella rufifrons Gray.

d' 2. Agageh Wells, Kordofan, November 1902.
Capt. Dunn tells me that this Gazelle does not range beyond
about 50 miles to the northward of El Obeid.

3. On a Collection of Tnrbellaria Polycladida from the Straits
of Malacca. (Skeat Expedition," lb99-1900.) By F. F.
Laidlaw, B.A. Cantab., Assistant Lecturer and Demon-
strator in tlie Owens College.^

[Received March 19, 1903.]

- (Plate XXIII. & Text-figures 50-56.)

The collection described below was made entirely by Mr. Evans
on the shores of a small islet called Pulau Bidan, a few miles noith
of Penang, It proves a most intei'esting one, and includes, so far
as I can discover, only one pi'eviously known species, Thysanozoon
auropunctatuvi Coll.

As in describing Mr. Gardiner's collection from the Maldives
[4], I have to note the scarcity or absence of the Eurylejotidae,
i-epiesented doubtfully only by a much damaged fi-agment in
Mr. Evans's seiies.

The only previous record of species from the shores of the
Malay Peninsula that I have been able to discover was made by
Collingwood, who desci'ibed and figured the following species from
Singapoie : —

Thysanozoon allmani.
Proceros hancockianus.

„ huskii.
Sphyngiceps lacteus.

Eurylepta kelaartU.
Elas'inodes ohtusam.
Leptoplana aurantiaca.

Grouping these specimens with those of the Skeat Collection and
adding Pseudoceros hedfordii fi'om Singapoi'e, kindly given me by
Mr. Lanchester, we get the following list classified according to
Lang's system [6] : —

Acotylea.

Plaxocerid.e.

* Planocera sp.
1.* Notoplana evansii, sp. n.

1 Communicated by Dr. S. F. Haemek, F.Z.S.

2 For explanation of the Plate, see p. 318.

302 MR. F. F. LAIDLAW O^S THE [-^W- 21,

Acotylea (con.).

LePTOPL ANID^ .

2.* Semonia penangensis, sp. n.

3.* Leptoplana malayana, sp. n.

4. „ ohtusitm (Coll.).

Cryptocelidid^.

5.* Bergendalia cmomala.
Latocestid^.

6.* Latocestus argus.
Cotylea.

PSEUBOCERID^.

7.

ThysanoZ'

oon alhnani.

8.*

11

auropunctatimi.

9.

Pseudocer

'OS hancockianus (Coll.).

10.

11

huskii (CoU.).

11.

11

kelaartii (Coll.).

12.

11

hedfordii, sp. n.

13.*

11

collingwoodii, sp. n.

14.*

Psettdocer

'OS ? ruhellus, sp. n.

DiPOSTHIDiE.

15.*

Asthenoceros woodivorfhi, sp. n.

Prosthiostomid^ .

16.*

Prosthiostomum 2}cdlidu7n, sp. n.

17.

11

aurantiacum (Coll.).

Species marked with an asterisk were collected by Mr. Evans.

In this eomniunication I have not ventured to give an account
of the anatomy of any of the species, contained in the collection,
which can be regarded as being at all complete. Siich an account
would have expanded the paper beyond limits reasonable in a
systematic description. I have attempted only to give such a
diagnosis of each species as shall render its future identification a
matter of tolerable certainty, and to call attention to any of the
more striking features which presented themselves.

Family PLANOCERiDiE.

Planocera sp. (Plate XXIII. fig. 1.)

A fragment of tissue of which I made microscopic sections
proves to belong to a species of this genus, but being only a
fragment I will not attempt to describe or name it, though it is
obviously new. The section passes through the penis and prostate,
and the spines lining the Ivimen of the f ormei- are of such a remark-
able character that I venture to figure one of them, in the hope
that a perfect example may be obtained ere long. The length of
each spine is roughly "08 mm., its breadth at the base "02 mm.

ISToTOPLANA, gen. nov.
NoTOPLANA EVANSii, sp. n. (Plate XXIII. fig. 2.)
A number of specimens belonging to this species were taken by

1903.] POLYtLADS OF THE " SKEAT EXPEDITIOX." 303

Mr. Evans on the shoi'es of the island of Pulau Bidiin, whei'e it is
evidently one of the commoner species.

The body of this worm is rather elongated, rounded in front
and with a pointed hindei- extremity. The arrangement of its
eye-spots is shown in the accompanying figure (text-fig. 50).

Text-fig. 50.

%."■••■

• •

Eye-spots of Notoplana evansii.

Unfoi'tiinately I have no notes as to coloui-, but, judging from the
spirit-specimens, the ci-eature is probably of a yellowish-grey
coloui-. with a small niunber of ii-i'egularly scattered black spots.
The gut-bi'anches aie numei'ous, and thei'e is no anastomosis.
The dimensions of an avei'age individual ai'e as follows : —

Length 25 mm.

Breadth 15 „

Tentacles f i-om ant. naargin 5 „

^louth-opening ,, 12 ,,

(5 . Aperture behind mouth 3*5 ,,

9. ,, behind male 1"5 ,,

Genital Apparatus (see text-fig. 51, 304).

The most stiiking featui'es of the male copulatoi-y appai-atus
ai'e the great length of the antral chambei- and the chambered
pi'ostate organ, the latter recalling the jii'ostate of certain
Leptoplanas, e. g. L. alcinoi. From the male aperture the antral
cavity (text-fig. 51, a.m.) extends forwards as far as the level of the
hinder end of the pharynx. It has very muscular walls and runs
sloping in an upward dii-ection. Lang, in his diagram of the
genital appaiatus of Hoploplana {Planocera insignis gi-oup [6]),
calls this chamber the penis-sheath (" Penisscheide), but this term
should, I think, be reserved for the sheath-like folds of the walls
of the anti-al chamber occurring in Cestoplaaa and in certain
Cotylean genei-a.

The lining of the walls of the antiul chamber in the pi'esent
species consists of a flattened ciliated (?) epithelium ; the muscular
wall consists of a thick layer of inteilacing circular fibres.

Proc. Zool. See— 1893, Vol. I. No. XX. 20

304

MR, F. F. LAIDLAW ON THE

[Apr. 21,

The penis (jo.) itself is small and projects into the antral chamber
at the upper anteiioi- end of the latter.

The penis is a fleshy organ, composed of longitudinal and
circular fibres which exhibit numerous nuclei (see Plate XXIII.
fig. 2). It is armed with a chitinous stylet.

The base of the penis is entei-ed by the short ductus ejaculatorius
{d.e.) which leads to the prostate region. It turns shortly after
leaving the penis and runs in a backward direction, so that a

Text-fig. 51.

a.m:

Male organs of Wotoplana evansii.
Tor explanation of lettering, see p. 318.

transverse section passing through the penis passed also through
the prostate (see PI. XXIII. fig. 2 and also text-fig. 51).

The prostate consists of a number, some nine in aU, of small
chambers {pr.c.) which lie about the ductus ejaculatorius and
open into it at their distal ends. The ductus, elsewhere with
muscular walls, here is lined only with a very flattened epithelium,,
outside which lies a second layer of similar cells belonging to the

1903.] POLYCLADS OF THE " SKEAT EXPEDITIOX." 305

inner walls of the pi-ostate chambers. Between each of these
chambers a fine double septum runs out from this inner wall (Plate
XXIII. fig. 2, s.). The prostate cells stain very feebly, and the
lumen of each chamber is well defined, and contains in most cases
a cei'tain amount of secreted mattei-. Outside tlie chambers is a
fairly thick layei' of circular muscle-fibres, and, l>e}-ond these, an
ill-defined layer of cells which appear to be glandular. Here
and thei-e are faint indications of pi-ocesses from these outer
glandulai- cells, piercing the muscle-layer of the prostate organ.
This muscle-layer is also ti-aversed by processes from the septa of the
chambei-s. Thei-e is no longitudinal muscle-coat to the pi-ostate.

Towards the level whei-e the prostate-chambers open into the
ductus ejaculatorius, some of them open into each other, so that
the number of chambers seen in cross-section is reduced to four
or five.

On the proximal, hinder, side of the prostate, the ductus is
continued back into the vesicula seminalis {v.s.), which is long and
contorted. It is lined with a ciliated epithelium, and its walls are
composed of i-egulai-ly airanged cii-culai- muscle-fibres. At its
exti-eme hindei- end it turns shai'ply forward and ends bUndly.
Just before it takes this turn forward, the two vasa deferentia
open into it, one on either side, aftei- piercing thi-ough its
muscular wall.

Thei-e is a spacious anti-um femininum suiTOiuided by the large
shell-glands, the secretions of which it receives. The vagina leaves
the anti'um dorsally, and i-uns forward for the first part of its
course thi'ough the shell-glands, the secretion of which it also
I'eceives. It then turns first doi'salwards and then backwards,
acquiring in its course a fine wail of circular muscle-fibi'es. After
I'unning back for a short distance, it receives the shoit common
duct from the two uteri. It then continues to I'un back as fai' as
the level of the antrum femininum, where it ends blindl}-.

It is evident that this species is closely allied to von Plehn's
Playiotata pi-omiscaa [7] ; in fact, the terminal parts of the male
apparatus in the two species have almost identically the same
structui'e.

It difiers from von Plehn's species in having the phar3'nx of the
noi-mal t}^e, not elongated transversely. This character is of
sufficient importance to justify the placing of these species in two
sepai'ate genera. In addition Notojjlana is characterised by
differences in the female apparatus, and by the gi-eat length of
the antral chamber of the male oi'gan.

The genus yotoplana may be defined, then, as follows : —

A Planoceroid genus ivlth styliform penis and unthout a bursa
copulatrix. The male antral chamber is very long ; there is a
complicated prostatic organ consisting of several chambers lying
aroiond the ductus ejaculatorius in front of the vesicula seminalis.
The latter is long and twisted. Body rather elongate, unthout
marginal eyes ; mouth- ojjening rather behind the middle ; pharynx
normal ; gut-branches mtmerous, ivithout anastomosis.

20*

306

MR. F. F. LAIDLAW ON THE

[Apr. 21,

fi'OBi the anterior end.

Family LeptoplaniDvE.
Leptoplana malayana, sp. n. (Plate XXIII. fig. 3.)

A number of specimens ; some, preserved in formol, are of a
uniform greyish-brown.

Length, about 35 mm.

Eyes, about 8 ,,

Mouth-opening, abou,t 15 ,, ,, ,,

c? aperture, about ... 10 ,, from the hinder end.

$ ,, ,, ... 1 ,, from male.

The arrangement of the eye-spots is that usual in the genus :
there are two small clusters of " tentacle- eyes " about 1*5 mm.
apart, and in front of these are a few scattei'ed spots on either side
of the middle line.

This species is closely allied to L. paciUcola von Plehn [7].

Genital Apparatus (see text-fig. 52).

The vasa deferentia (v.cZ.) open into the proximal (anterior) end
of the vesicula seminalis (v.s.) ; the latter, of considerable length
and much twisted, is lined with ciliated epithelium, and has thick

Text-fis. 52.

v.s.

sh.^l

ace. ves.

Genital apparatus of Leptoplana malayana.

a.f., antrum femininum ; acc.ves., accessory vesicle ; sh.gl., sliell-glands ;

u.t., uterine duct ; v.d., vasa deferentia. For other letters see p. 318.

muscular walls made up of circular fibres. At its distal end the
vesicula narrows into the ductus ejaculatorius (d.e.), which is
shorter, but, like the vesicula, much twisted. Its walls are like
those of the vesicula, but it is much narrower and has a less well-
developed muscle-layer. The penis is small, fleshy, and unarmed,
and projects into the large antral chamber, which has thick mus-
cular walls and extends vertically upwards almost to the level of
the dorsal body-wall muscles. (See Plate XXIII. fig. 3.)

The female opening lies close behind the male, and leads into a
moderately large antral chamber {a.f.) with musculai- walls. Into
this the vagina opens from above. This latter organ is a long and
tortuous duct of varying diameter ; the shell-glands (sh.gl.) open
into it through nearly its whole length. At its hinder end, near

1903.] POLYCLADS OF THE " SKEAT EXPEDITION." 307

the point wJiere tlie uteri open into it, the muscuhir walls, which
are elsewhere feebly developed along its course, become thicker,
but not very mnrkedly so (cf. L. caUfoniica von Plehn). The
two uteri i^at.) unite as they enter it from below, and behind them
there is a veiy small accessoiy vesicle (acc.ves.) with muscular
walls.

The genus Leptoplana, as at present constituted, contains a
consideiable number of species, many of them only i-eferred by
Lang with doubt to the genus. In the majority of the species,
the anatomy of which has been investigated by means of sei'ial
sections, the genital ajipai-atus pi-esents the following chai-acters : —

The penis is directed backwai-ds. The vasa deferentia open
into a muscular vesicula seminalis ; from this the ductus ejacula-
torius I'uns backwai-ds to open into a prostate organ, -which maybe
chambered ; le;xving this, the duct runs back into the penis,
which may be armed with a chitinous stylet.

The vagina runs back from the antrum femininum, receives the
secretion of the shell-glands, and further l)ack the two utei-i open
into it ventrally, usually by a connnon duct. An accessory vesicle
is generally present. The most familiar species of the genus, L.
tremellaris, differs from the majoiity of species in the structure of
its male apparatus sufficiently to permit us to put it on one side
to foi-m of itself a section of the genus fui-thei- characterised hy
the possession of a ventral sucker between the genital openings.
Anothei' species, L. suhviridis von Plehn [8] ( = X. 2^<^''''^dalis mihi
[4]), approaches Biscocelis tigrina in the structure of its female
oi-gans, and accoixlingly we may put this species also in a section
apart.

The remauiing species I have attempted to classify beloA\-, so f;ir
as is at present possilale, according to the sti-ucture of the penis
and prostate. I trust that I will not )>e thought to have laid too
much stress on the structui'e of these organs in dealing with this or
other genera. It seems to me that, of all the characters that
present themselves for classifying this order of Turbellaria, these
ai-e the most useful, and that they are as reliable as any of the
other chai'actei-s employed foi' this purpose, such :is structui-e of
the pharynx, number and arrangement of eye-spots, presence or
absence of tentacles, &c. &c.

I propose, then, to group the species in two sections. A, and B,
so fai- as oui- knowledge of theii- anatomy pei-mits : —

A. Penis provided with a. stylet.

(a) Prostate complicated by i-adially arrangetl "Drusen-
kanUle."

1. L. kukenthali v. Plehn. Spitzbergen [7].

(6) Prostate dividerl into chambers lying parallel to the
ductus ejaculatorius.

2. L. vitrea L:nig. Mediterranean [61.

3. L. alcinoi Schmidt. „ [61.

308 MR. F. F. LAIDLAW ON THE [Api'. 21,

(c) Prostate not chambered.

4. L. pcmamensis v. Plehn. Gulf of Panama [7].

5. L. californica v. Plehn. California, Chatham Is. [11]

6. L. nationalis v. Plehn. Ascension Is. [8].

The following also belong to section A, but their position in
the section cannot be determined without information as to the
prostate characters.

L. variabilis Giard. 1

L. ellipsoicles Yerrill. I New England [13].

L. virilis Yerrill. J

L. drmhachensis Oersted. Norway [6].

L.fallax (de Quatrefages) [6] should perhaps form a separate
section of the genus on account of the great length of its stylet.
It occurs on the south side of the English Channel.

B. Penis unarmed.

a. Prostate chambered.

7. L. chierchcB von Plehn. Callao [7].

/3. Prostate not chambered.

8. L. pallida (de Quatrefages). Mediterranean [6].

9. sp. (unnamed). Maldives [4].
y. No definite prostate.

10. L. jmcijicola von Plehn. Valparaiso [7].

11. Z. malayana, sp. n. Straits of Malacca.
Belonging to this section but of uncertain position in it is

L. ancjtbsta Verrill [13] from New England. L. lacteoalba Yerrill
[14] is insufficiently characterised, but said to be like L. pallida.

Semonia penangensis, sp. n.

Two specimens from Pulau Bidan. No note available as to
colour. This interesting species agrees fairly closely with the
only other species of the genus at present known, viz., S. inacu-
lata von Plehn, from Java [7]. It is, however, readily distinguished
from that species by its external characters, the arrangement of
the eye-spots being obviously very diiferent and showing some
approximation to the condition found in Discocelis tigrina (see
text-fig. 53, p. 309).

The dimensions of the larger specimen are as follows : —

Total length, about 22 mm.

Breadth, about 11 „

Mouth-opening, about ... 9 „ from the hind end.

Genital aperture 1 „ behind mouth-opening.

Brain, about 4 „ from the anterior margin.

The marginal eyes extend back for abou:t 8 mm. on either side.
The arrangement of the brain and tentacle-eyes is shown in
text-fig. 53, p. 309.

An important distinction between this species and Semonia
macidata is afibrded by the fact that in the latter the testes and

1903.] POLYCLADS OF THE " SKEAT EXPEDITION." 309

ovaries lie on the dorsal side of the Ijody (see von Plehn [7]),
whilst in the present species the position of the testes is normal.
i. e. on the ventral side.

In one or two respects this species approaches the closely allied
Discocelis tiyrina, and is pei'haps intermediate between that species
and *S'. maculaia. But the latter s})ecies is also certainly closely
jillied to Discocelis, and I am rather surprised to find that von
Plehn has not instituted any compaiison between the two genera.
/S. penanyensis difiers fi'om S. maculaia, and approaches Discocelis
tigrina, not only in the features already mentioned, viz., the
arrangement of the eye-spots and the position of the testes, bnt
also in that the utei'i unite to enter the vagina by a shoi't common
duct, and in liaving a veiy V)hnit, almost square penis.

Text-fie-. 53.

'%
o

" Braiii-eyes " of Semonia penangensis.

It difiers from Discocelis, and resembles S. maculaia, in being
without the characteristic large prostatic cells which occur in the
epithelium of the penis and of the antrum, and without the
remarkable paired structiu'es which run foi'wai-d fi'om the accessory
vesicle oi Discocelis tigrina. Further, like S. maculata, it possesses
a definite vesicula seminalis, which appears to be absent in
Discocelis.

Lastly, *S'. penangensis is without the cuiious glandular vesicles
found along the anterioi- ends of the vasa deferentia in S. maculaia ;
and the vagina is prolonged, beyond the point whei'e it I'eceives
the openings of the \iteri, into a veiy small accessory vesicle — so
small, in fact, that it almost escaped observation. This may be
due to the fact that the only specimen available for section-cutting
was not quite mature; oi- moi-e probably, since the uteri were full
of apparently ripe eggs, that tliis organ is undei'going degeneration.

It is obvious that the distinction between Discocelis and Semonia
is a very slender one, but still, I think, sufiicient to wari-ant the
retention of the latter as a valid genus. The most important
chai-acters separating the two genera are the absence of a vesicida
seminalis in Discocelis, and the curious hoi-seshoe shape of the
accessory vesicle in that genus.

310 MR. F, F. LAIDLAW ON THE [Apr. 21,

Family Oryptocelidid^.

Bergendalia anomala, gen. et sp. nov. (Plate XXIII. figs. 4,
5, 6, & 8.)

A most remarkable and interesting form, probably allied to
the anomalous genera Gryptocelides and Polypostia, desci"ibed by
Bergendal [1], and provisionally referred to the same family with
them.

The structvire of the female tei-minal ducts is, so far as I know,
unique, and approached only b}-- the sj)ecies referred to Trigono-
poms and Polyporus (see von Plehn [11]).

Only a single specimen was obtained. It is rather a large
form, with a total length of about 60 mm. and breadth about
27 mm. The mouth- oj)ening is some 15 mm. from the hind end,
and the genital pores lie halfway between these two points.

The margin of the body is complete, surrouiaded by a continuous
rather dense row of eye-spots. Thei"e are none of these, apparently,
over the brain.

Ooloui- in the spirit-specimen uniformly greyish white (see
PI. XXIII. fig. 8). The specimen appears to be in an early
stage of sexual maturity, since no trace of ovaries or testes can be
discovered.

The pharynx is large and much folded, the gut-branches are
numerous and anastomose freely.

The cells of the epidei-mis are elongated, especially on the dorsal
surface. True rhabdites are absent, but in place of them the
epidermal cells are crowded with pseudorhabdites which are of a
coarsely granular texture, faintly stained and columnar in shape.
In my sections (stained with Grenadier's htematoxylin) certain
gland-cells lying within the muscle-layers of the body-wall,
especially on the ventral side, are deeply stained ; from these cells
I'un processes Avhich piei'ce the muscle-layers and basement-
mLembrane, and make their way to the surface through the
epidei-mal cells. These deeply-lying gland-cells and their processes
have rather a spongy appearance, due to their preserA'ation not
being quite perfect.

Genital Organs (text-fig. 54, p. 311).

The penis (p.) is a small fleshy organ composed of nucleated
longitudinal fibres (PL XXIII. fig. 4). Its outer side is lined with
cells continuous with those lining the antrum masculinum, but
whereas the latter are ciliated, those coveiing the penis are non-
ciliated. The base of the penis is pierced by a duct rvmning
nearly vertically upwai-ds. Immediately after leaving the penis
this is joined by two small ducts — the vasa deferentia {y.d^ —
which run forward on either side of the middle line. In one of
them, at the level of the hinder end of the pharynx, there is a
slight dilatation containing spei-matozoa.

The vesicula (j'Jr.) lying above the penis is small. Its wall

1903.] POLYCLADS OF TUE " SKEAT EXPEDITION." 311

consists of a very tluii l;iyer of circular nmscle-fibres. lineil with
a cubical epithelium, the nuclei of the cells of which lie close to
the lumen, and which has rather the appearance of an exhausted
secretory tissue. This oi'gan I call the prostate (PI. XXIII. fig. 4,
pro.). About halfway between the penis and fenxale aperture lies
an oi-gan which beai-s some i-esemblance to the penis already
described. This organ constitutes one of the most interesting
features of this anomalous species. It consists of a small pro-
trusible penis-like organ (7;.?), about one-third the size of that
first described, lying in a small antrum. It is pierced by a short
duct, which luns into it from a small non-muscular vesicle which
lies immediately above it (jjr.f). This small vesicle is of about
the same size as the prostate. Thei'e is no communication, so far
as I can discover, between the duct connecting these structures
and the va.sa deferentia (PI. XXIII. fig. 5).

ace.

Geiiitnl apparatus of Bergendalia anomala.

ace, accessor}' part of vagina ; p. ?. penis-like organ ; pr. ?, prostate-like organ ;
s., spiral coils of vagina. For other letters, see p. 318.

I can suggest only two explanations of the presence of this
second penis-like organ and its accessories.

Fii-stly, that it may be regarded as a pi-ostatic sti'uctiu'e ^^"hich
has lost its connection with the penis and developed an intro-
mittent terminal part of its o^\•n. Such a state of afiairs is found
in the Ootylean family Diposthiidte ; but in that case both penis
and pi-ostate open into a common antrum, and there is what
appears to be a pi'ostate gland in connection with the penis in
Bergendalia, a fact which makes sti-ongly against this view.

The second possible explanation is that the oi-gans under dis-
cussion ai'e the vestiges of a second penis. The fact that they l^ear
a close resemblance to the functional penis tends to support this
view, as does also, I think, a comparison with Cryjitocelides of
Bergendal [1].

This is the only other Polyclad described, .so far a.s I know, in
which two penial organs lie behind each other on the middle line;
but, according to Bergendal, they lie behind the female aperture,
and, further, lioth open into a common antrum, whilst in some cases

312 Ma. F. F. LAIDLAW ON THE [Apr. 21,

thei'e may be four or even six penes present. In another genus
presumably allied to this, viz. Polypostia Bei-gendal [1], the penes
lie in large numbei'S around the female apei'ture.

The female apparatus is no less remai'kable than that of the male.
The vagina {va}) runs forwai-d for some little distance from the
apertiu'e, then turns upwards. As it does so, it is twisted into
a remarkable spiral coil, making some five complete turns. . It
then runs backwards, narrows considerably, and soon receives the
openings of the two uteri {ut.) on its ventral side. Beyond this
point it is continued back as a narrow accessory vesicle {accves.)
about as far as the level of the female aperture, when it tui-ns
shai-ply ventralwards and oj^ens to the exterior hy the caitriwi.

The only other Polyclads with which I am acquainted that
possess a secondary female opening ai'e Trigonoporus of Lang [6]
and Polyporus of von Plehn [11], but in both these cases the
second opening is quite distinct from the primary one (the true
antrum femininum).

The curious spiral twisting of the vagina in the present species
is, so far as I know, unparalleled in the oi-der.

For the first part of its coui'se, %. e. whilst it is running forwards,
the vagina is lined with elongated cokimnar ciliated cells, the nuclei
of which lie near their bases. Outside this epithelium is a thin layer
of circular muscle-fibres, and beyond these, in my sections, can be
seen a great number of nuclei massed round the vagina, and
probably belonging to gland-cells. Where the vagina turns
dox'salwards and becomes spirally twisted it has narrowed slightly,
but its epithelium i-etains the characters already mentioned. The
muscle-sheath does not follow the individual folds of the spiral, but
forms a continuous covei-ing for that part of the duct (PI. XXIII.
fig. 6, m.s.). The rest of the terminal female ducts are precisely
similar in character to the first part of the vagina, only narrower.
There seem to be no special shell-glands present.

The genus may be defined as an Acotylean genus in lohich behind
the functional penis a small second penial organ occurs. The
accessory vesicle of the female apparatiis opens to the exterior
through the antrum femininum. Body pointed at either end.
Marginal eye-spots present ; moibth subcentral.

Family Latocestid^.

LaTOCESTUS ARGUS, sp. n.

Four specimens were collected by Mr. Evans. One of these is
labelled " Chocolate-brown above, slightly lighter below."

The largest specimen has a total length of about 30 mm. and a
breadth of 5 mm. The mouth is 2*5 mm. from the hinder end of
the body.

This species is proportionately longer than the other member of
the genus which I have had an opportunity of examining. Its
most striking feature is the presence of a crowded row of eye-
spots running completely round the margin of the body. The

rOLYCLADS OF THE " SKEAT EXPEDITION'.

313

1903.]

arrangement of the eyes at the anterior end of the body is shown
in text-fig. 55 A and towards the hinder end in 55 B.

The only other elongate form exhibiting this chai'acter with
which I am acquainted is CestO'plana ? maldivensis niihi. This is
ahnost certainly a Latocestus.

Text-fig. 55.

Eye-spots of Latocestvs aryus.

I can detect no rhabdites in the epidermis. The parenchyma
is very dense, and nervous tissue abundant on the ventral side
just within the body-wall muscles.

The terminal parts of the reproductive apparatus are almost
exactly similar to those figured by von Plehn [7] for Latocestus
paciAczis.

314 MR. F, F. LAIDLAW ON THE [Apr. 21,

As in that species, the testes as well as the ovaries lie on the
dorsal side of the body.

Thysaxozoon auropunctatum Coll.

Thysanozoon aibropitnctatum Lang [6] ; von Stummer-Traiinf els
[13].

One specimen, unfortunately immature and in a poor state of
preservation, although it has retained its colour fairly well.

PsEUDOCEROS BEDFORDi, sp. n. (Plate XXIII. fig. 9.)

This strikingly handsome and large sjDecies was obtained by
Messrs. Lanchester and Bedford in Singapore Hai-bovir. The
single specimen, kindly given me by Mr. Lanchester, was unfortun-
ately in fragments and fully one-half of the specimen missing, so
that a satisfactoi-y examination was not possible. I believe, how-
ever, that it was provided with a pair of penes. The colouring is
very beautiful, and quite suificiently marked to render the identifi-
cation of this species in the future a simple matter, thanks to the
careful drawing for which I am indebted to Miss Dust.

On the label accompanying the specimen is written " Singapore
Harbour, from tide-marks to 10 faths." I venture to associate
with this species the name of my lamented friend the late
Mr. Bedford.

PsEUDOCEROS COLLINGWOODI, Sp. n.

Closely allied to Ps. cerebralis (Kelaart) and Ps. striatus
(Kelaart), but smaller than either. It has the same i-egular
convolutions at the margin of the body that are shown in
Collingwood's figures, and the coloration is similar especially to
Ps. cerehi'alis.

Length, about 30 mm.

Breadth ,, 25 ,,

Mouth-opening, about 3 ,, from the anterior margin.

Sucker about 8 ,, behind the mouth.

The specimen is immature and the penis is unpaired. Colour —
dorsal surface mottled dark brown and brownish white. The
extreme margin is white, but the white rim is exceedingly thin ;
just inside this is an equally fine black line. Ventral surface pale
brownish white, becoming darker towards the margin, which is
edged with black and white just as on the dorsal surface.

PsEUDOCEROS ? RUBELLus, sp. n. (Plate XXIII. fig. 10.)

A number of specimens of this very small species were collected.
One of these was " found under a stone between tide-marks. It
was magenta-red in colour. Xov. 1899." Though very small,
these specimens appeal- foi- the most pai't to be mature. They
bear a close resemblance to Pseudoceros kelaartii (Coll.) in colour,
but the lattei- species is much larger and has the eye-spots arranged
quite difi^erently. The arrangement in the present species is shown

1903.] POL\'CLADS OF THE ''SKEAT EXPEDITION." 315

in fig. 10 of Plate XXIII. The tentacles also in Colliugwood's
species ai-e nincli moi'e prominent. The penis is unpaired. The
body is nearly circular and the sucker median. Total length
about 5 mm., bi'eadth 4 mm. The ai-i-angement of the eye-spots
is veiy diflfei-ent from that normally found in Pseiuloceros, but I
cannot find any other charactei-s distinguishing it from that genus.

At least two othei- species of this genus ai-e repi-esented in the
collection. One of these is a very small individual of the
F. superb us- gvouTp (perhaps P. hancockianus) ; whilst several
specimens belong to a species coloui'ed exactly as Proceros con-
cinnum Coll., viz., they are blue, with orange mai-gin and median
stripe. They are, however, very much smaller, about 8 mm. to
10 mm. in length, and of a different shape, not pointed at their
hinder end, and have small, folded tentacles, whereas Collingwood's
species has pointed tentacles. Proceros concinnum in fact is almost
certainly a Euryleptid, whilst these specimens are undoubtedly
members of the genus Pseudoceros.

I prefer, howevei', not to name them at present.

Family Diposthiid.e.

AsTHEXOCEROS, gen. nov.

AsTHENocEROS wooDWORTHi, sp. 11. (Plate XXIII. fig. 7.)

Two specimens, without notes as to colour ; but, to judge from
these spirit-specimens, the species is of a leddish-brown colour,
darkest on the middle line.

Length, about ... 17 mm.

Breadth U „

Mouth-opening... 7 ,, from the anterior margin.

Sucker 1*5 ,, behind mouth.

The male and female apertures lie in the usual oi-dei- between
the "mouth" and sucker.

The pharynx is large and much folded, its opening being sub-
median. This featui'e will serve at a glance to distinguish this
species from any of the Pseudoceridfe. In describing the type of
the family, Woodwoith [16] makes no statement as to the pharynx,
but, to judge from the figure given, it is median oi' snbmedian as
in the present species.

Tlie bod}^ is flat and almost circular ; the anterioi- mai'gin is
feebly folded, and thei'e is a small group of eye-spots on either
side of the middle line, but there are no very definite tentacles.
The prostate body lies in front of the jjenis.

In this respect Asthenoceros diffeis fiom the type-genus Dipos-
thus. There ai-e no biain-eyes.

Only one of the two specimens was cut into sections, tx^ans-
versely ; the other specimen appesirs, from an examination of it
when cleared in oil of cloves, to be quite immature. The speci-
men serially cut presents certain peculiar featuies, which I am at

316 MR. F. F, LAIDLAW ON THE [Apr. 21,

a loss to explain satisfactorily. In the first place, the testes and
ovaries are very immature, and appai'ently only the terminal parts
of the genital ducts are developed. On the othei- hand, there is,
immediately over the penis, a large mass of mature spermatozoa
lying in a chambei- the character of which cannot be determined
from my specimens, since it appears in places to have indications of
a proper lining-epithelium of its own, and again in places seems to
be merely a gap in the parenchymatous tissue. There are very
faint indications of a duct running from this chamber down in
the dii-ection of the penis, but this duct cannot be traced far.
Woodwoi-th [16] suggests in connection with the specimens of
Diposthus described by him, that they were in a late stage of
sexual activity, and that in consequence of this the sexual organs
were in a reduced condition. Possibly in my specimens of
Asthenoceros a similar state of afikirs occurs, but, as I have
already stated, testes and ovaries in a very immature sta,te are
present. There is also a possibility that the ripe spermatozoa
may be derived from another individual by hypodermic injection ;
but I do not think this is the case — firstly, because the penis is
not armed with a stylet ; and, secondly, because the spermatozoa
lie over the penis, and because, as already stated, there are traces
of a duct running towards the penis from the chamber in which
they lie.

The terminal parts of the male apparatus histologically resemble
those of BipostMts. The prostate and penis are separated into
two distinct organs, " both of which are doubtless intromittent,"
both opening by a single gonopore. The prostate lies immediately
in front, the penis directly over the aperture. It is rather
feebly supplied with muscle-fibres, and the secretory cells lie in
the middle of the organ, but there is no lumen apparent
(cf. DipostMis corcdlicola).

The penis is much more muscular, and on the outer side has a
very definite series of cii'cular and longitudinal fibres. Nuclei in
it too are much more niunei-ous (PI. XXIII. fig. 7).

In connection with the female organs there are traces of two
pairs of viterine vesicles. The uteri are very small and difficult to
distinguish from the surrounding tissue, and the ducts running
to them are merely solid rods of cells of an embryonic appearance.
The uteri open into the hinder end of the vagina. The antrum
is deep, and widens at its upper end where it receives the secretion
of the shell-glands.

The body epithelium is very densely crowded with small
I'habdites, and on the dorsal side especially with pseudoi-habdites.
Scattered through the parenchyma, more particularly in the region
of the sucker, are numbers of small rounded darkly-staining cells
the nature of which is doubtful. The ventral surface of the body
projects immediately behind the pharynx into a prominent median
ridge which carries the gonopores and the sucker ; behind the
sucker the ridge disappears.

I have given as complete an account as possible of the cha-

1903.] POLYCLADS OF THE " SKEAT EXPEDITION." 317

racters of the genital organs, but, owing to the comlition of tlie
specimen, it is obviously fai- from being satisfactory, and in order
to deal fully with this interesting species more material is
necessary.

Prosthiostomum pallidum, sp. n.

One specimen, from tlie sea-shore, Dec. 1899.

Total length, about 20 mm.

Breadth, about 4 ,,

Anterior margin I'ounded. Arinxngement of eye-spots shown
in text-fig. 5G.

Text-fio-. 56.

Ej'e-spots of Prosthiostomum pallidum.
e, margiual ; br.e, bi-aiii ej'e-spots.

Closely allied to P. siphancuhis of the Meditei-i-anean by the
ari-angement of the eye-spots, and agreeing with it apparently in
being, so far as the spirit-specimen shows, of an uninterrupted
dull grey-bro-wTL colour, it differs sufficiently in that the two rows
of brain-eyes diverge continuously from one another from before
backwai-ds, whilst in the Mediterranean species these two rows
converge at their middle.

It is also readily distinguishable, I think, from the latter species
by its smaller size. The single specimen obtained by Mr. Evans is,
to judge from the state of the sexual apparatus, fully mature.
P. siphitncidus, moreovei', has the genital sexual organs relatively
much smaller, to judge from Lang's figui'es (see Lang, ' Poly-
cladida,' pi. 5. fig. 3).

P. pccUidicm Ls certainly veiy distinct from any species from
the Indian Ocean that I have had the opportunity of studying,
and also, I believe, from the species found in the Pacific.

I believe that Leptoplana aii,rantiaca of Oollingwood is really a
Pi-ostkiostomum. It has the shape and 2)roportions of a member
of that genus, whilst its eye-spots have the characteristic Pros-
thiostomuiii arrangement. It may, of course, be identical with the

318 ON THE POLYCLADS OP THE " SKEAT EXPEDITION." [Apr, 21,

species described above, but, judging from Colling-wood's figui-e,
in the arrangement of the eye-spots it is sufficiently distinct.

Liter atttre.

1. Bergendal. "Kongl. Fysiogr. Sallskapet," Lund Handlingar,

Ny Foljd, ] 892-93, Bd. 4.

2. CoLLiNGWooD. Trans. Linn. Soc. Lond. ser. 2, Zool. i. 1875,

p. 83, pis. 17-19.

3. Graff, von. Zeitschr. wiss. Zool. Iv. 1893, p. 189.

4. Laidlaw. Fauna and Flora of the Maldives and Laccadives,

i. p. 282, pis. xiv., xv.

5. Mem. & Proc. Lit. &, Phil. Soc. Manchester, xlvii. p. 1

(1903).

6. Lang. Naples Monogr. xi., 1884.

7. Plehn, von. Jena. Zeitschr. xxx. p. 137, t. viii. & xiii.

8. Ergebn. Plankt.-Exp., Heft i., 1896.

9. Semon's Zoologische Forschungsreiseil, Bd. v. pp. 329-

334, t. xxiii.

10. Abh. Senckenb. Gesell. xxiv. ii. pp. 145-146 (1897).

11. Jena. Zeitschr. xxxi. p. 90, t. v. (1898).

12. Zool. Jahrb. Syst. xii. pp. 448-452 (1899).

13. Stummer-Traunpels, von. Zeitschr. wiss. Zool. Ix. p. 688,

t. xxxv.-xxxviii.

14. Verrill. Trans. Connect. Acad. viii. p. 459, pi. xliv.

15. ,, „ ix. p. 523.

16. Woodworth. Bull. Mus. Oomp. Anat. xxxii. 4, pp. 63-67,

pi. i. (1897).

EXPLANATION OF PLATE XXIIL

Fig. 1. Spine from the lumen of the penis of Flanocera sp., p. 302.

2. Section passing through the penis and prostate organ of NotoplMm evansii,

p. 302.

3. Transverse section through the penis oi Leptojplana malayana, p. 306.

4. Transverse section througli the penis and prostate gland (?) of Bergeiidalia

anomala, p. 310.

5. Section through the rudimentary organ of penis-like structure of the same

species, p. 311.

6. Section through the spiral part of the vagina of the same, p. 312.

7. Section across the antrum masculinum of Asthenoceros tooodivorthi, p. 315.

These organs are directed obliquely forward, and consequently both appear in
some of the sections. The granular appearance of the prostate in this region
is rather exaggerated in the figure, but it becomes more marked towards its
exti'emitj'.

8. Sergendalia anomala, p. 306.

9. Pseudoceros bedfordi, fragment of anterior part of body, p. 314.
10. Pseudoceros ? rubellus, p. 314.

Explanation of Lettering of Plate and Text figures.

a.m., antrum masculinum.
a.m.l., muscular walls of antrum mascu-
linum.
d.e., ductus ejaculatorius.
d.pr., prostate duct.
ep., epidermis.

m.pr., muscular walls of prostate. | t., tentacles.

pr.c, chamber of the prostate.
pro., prostate.
ppr., prostate-like organ.

s., septum between chamber of

prostate.
St., stylet of penis.

m.v., muscular sheath of vagina.

p., penis.
pp., vestigial penis !
pr., prostatic cells.

va., lumen of vagina.
v.d., vasa deferentia.
V.S., vesicula seminalis.

190.'].] ox THE TUAXSrOSITIOX OF MAMMALIAN TESTES. 319

4. On the Phyloo-enetic Cnnso of tlic Transposition oF tlie
Testes in Mammal la : with Remarks on the Evolution
o£ the Diaphraom and the Metanephrie Kidney. By
W. Woodland, University College, London.'

[Received March 3, 1903.]
(Text- figure 57.)

Tlie sulijtK;t of th(! descent of the testes in ]\riimiu:di!i, notwitli-
standing its biononiic intori'st, has liitherto rt'ceived but littk^ atten-
tion from the phylogenetic standpoint ; indeeil, so far as 1 know,
its intei'protation has not hitherto been attempted. To exhibit
the general nature of the problem and the apparent dilHculty
attending its solution, tlie following paragraph from Mr. Spencer's
'Principles of Biology' (vol. i. p. 573) may be quoted : — "But
now let it be confessed that though all phenomena of organic
evolution nuist fall within the lines above indicated, there remain
many unsolved problems. Take as an instance the descent of the
testes in the Mammalia. Neither direct nor indirect equilibration
accounts for this. We cannot consider it an adaptive cliange,
since thei'e seems no way in which the production of sperm-cells,
internally carried on in a bird, is made external Vjy adjustment to
the changed requii-ements of mammalian life. ISTor can we ascribe
it to survival of the fittest ; for it is incredible that any mammal
was ever advantaged in the struggle for life by this changed
position of these organs. Contrariwise, the removal of them
from a place of safety to a place of danger would seem to be
negatived by natural selection. Nor can we regaixl the tiuns-
position as a concomitant of re-equilibration ; since it can hardly
be due to some change in the general physiological balance." I
agree with Mr. Spencer that the descent of the testes can neither
have been a change adaptive in nature, noi- a result of the
operation of natural selection ; on the other hand, I must dis-
agree with the statement that the phenomenon cannot be
regarded " as a concomitant of re-equilibivition," since, as will be
shown, I attribute it to the direct action of the conditions of life
found in the Mammalia.

To ensure due appreciation of the significance of the following
statements, it will be as well to here indicate the nature of the
theory about to be advanced. Of possible causes efi"ecting the
transposition of the testes, there are two categories : (a) causes
which as regards their eftects on the oi'ganism mediately
or immediately impart advantage, so supplying the requisite
material for natural selection ; and (6) causes which in their
eflfects on organisation bear no appreciable relation to either
advantage or disadvantage. Since those of the former categoiy
are inconceivable in this connection, we are led to conclude

1 Communicated by Prof. E. A. Minchin, F.Z.S.
Pkoc. Zool. Soc— 1903, Vol. I. No. XXI. 21

320 MR. W. WOODLAND ON THE [Apr. 21,

that the transposition is an inevitable concomitant of some other
constant feature of the animal's existence, thus not having arisen
in relation to ulterior ends. Again, since the testes consist of
ordinary matter possessing inertia and mass, their translation
(involving rupture or distension of the mesorchium) implies
mechanical force. From both of the foregoing considerations,
and from the fact that no other efficient cause is imaginable,
it is probable that the displacement results from the reaction
en the part of the testes to the incident forces, which, it can
be shown, are generated by bodily activity under mammalian
conditions of life. The theory here advocated is to the effect
that the descent of the testes in the Mammalia has been pro-
duced by the action of mechanical strains causing rupture of the
mesorchial attachments, such strains being due to the inertia of
the organs reacting to the impulsiveness involved in the activity
of the animals composing the group.

Mechanical Aspects of Organisation.

Before entering upon a discussion as to the production of the
forces above mentioned and the manner in which they have
acted, it is well briefly to outline the entire argument from the
mechanical standpoint. In the first place, accelerations imparted
to the body as a whole generate strains (or stresses) in connection
with the attachment or other means of support of every com-
ponent viscus, and the degree of such accelerations (and therefore
of the strains and stresses) is obviously dependent upon the
reaction which occurs between the animal's body and the medium
in or substratum on which it is supported, being directly pro-
portional to the product of the powers of resistance possessed by
the substances constituting the same. Secondly, it must be
pointed out that the accelerations to which we refer, and to
which alone we attribute any importance, are those involved in
the impulses communicated to the body during the actual con-
tinuance of locomotion, and which inevitably result fiom the
mode of action of the propelling agency, whatever its nature.
The ordinary non-impulsive accelerations involved in the starting
of an animal into motion from a state of rest, or its converse, do
not concern us. In considering, however, the arboreal habits of
e.g. Primates, the case is different, accelerations involved in
motion from rest here being decidedly impulsive in nature. Now,
among the Vertebrata, it is evident that fishes, aquatic amphibia,
and birds severally exist in media which, owing to their mobility,
negative the occurrence of accelerations of high degree, i. e.
impulsiveness. Hence, as regards these accelerations, we have
to consider only terrestrial animals, which come into contact with
a substratum possessing sufiicient power of resistance to afford a
reaction of mailced intensity. Of these terrestrial animals the
mammaliaii group is at once distinguished from terrestrial
Amphibia and Reptilia, both by the high degree of activity which

1903.] TUAXSrOSITION' OF MAJrJIALIAN TESTKl=!. 321

its members exhibit, and by the moi'c pci'fect adfiptatiou of tlic
body (relation of limbs to trunk, &c.) to the locomotion in which
the superior activity is mainly manifested. Consequently, in
mammals, the reactions which occur between the organism and
the substratum vastly exceed in magnitude those occurring in
lower gi'oups, and we have in this distinction a factor which in
considering bodily confoiniation is worthy of all attention.
Since in mammals tlieorgnnisMtion is subject to such considerahle
forces, it follows that if stal)ility of position Ijc required in con-
nection with organs possessing jipjjreciaLle mass, i. e. if these
organs are to adjust themseh es to those transposing inlluences
which arise out of inq^idsiveness, fixative structures must bo
developed in response to the demand. For, as remarked above,
a sudilen accelei-ation imparted to the body is necessarily trans-
mitteil to an attached organ through its attncliment, which, if the
organ be of consideiublc mass, is thereby su1)jected to an intense
strain, possibly ending in disruption. Stability of position of an
organ may be necessitated either on mechanical or physiological
grounds : thus, apart from considerations of function, viscera of
large mass (stomach, liver, intestine) must conform, as regards
situation in the body-cavity, to the principle that, cceteris paribus,
the more anterior the centre of gravity of the body the greater
the facility of progression ^ ; and again, from the functional
standpoint, the kidneys must in active animals be maintained
anteriorly in order to ensure j)roximity to a vigorous blood-suj)ply.
It will be evident from this, that if the position of an organ be of
no economic moment — if thei'e be no need of localisation either on
account of mass or volume for preservation of bodily equilibi'ium,
or on account of nature of function for j)hysiological etticiency, —
retentive structures will not tend to be formed, and the organ, if
of sufficient mass, will lespond by transposition to any consideralile
force incident upon it — a trait which in the course of generations
will be inherited.

I shall no^v proceed to examine more in detail the production of
the forces hitherto assumed to be capable of eflecting the trans-
position of the testes— to enquire whether the intensity and
direction of these forces are such as are congruous with the
hypothesis advocated.

Mammalian Locomotion.

In view of the importance that is here attached to the character
of mammalian locomotion, it is needful to discuss the sidjjcct in
detail. Every unsupported mass descends to the earth with a
given acceleration, and the magnitude of tlie force required to
re-elevate the body to its initial position is directly propoi-tional

1 Well shown by measurements of the position of the centre of gravity in Pisces,
and by the abdominal contours of active cursorial mammals as compared with those
of species more slothful in habit (compare e.^. a greyhound or race-horse with a
St. Bernard or cart-horse, or, more generally, lierbivores witli ciirnivores, though
the dlflerent capacities of the digestive organs have hcix- to be ooiisid;roil).

21*

322 MR. W. WOODLAND ox THE [Apr. 21,

to the height tliroiigh which upheaval is effected, and inversely
proportional to the period of time occupied by upheavaL This
statement in its application to terrestrial locomotion provides
the clue for the solution of the present problem. With the
exception of aquatic organisms, which exist in a medium so
similar in density to themselves that the influence of gravity is
not felt, all animals which fly, crawl, leap, or run are subject to
this primary condition of self-support imposed by the earth's
attraction. In Aves the body is not entirely unsupported, the
resistance of the air to the large expanse of body yielding con-
siderable aid in this respect, and the small amount of self-
sustainment needed can be provided by the minor elevations
imparted by individual strokes of the wing. The reactions
between the wing and the air are of necessity small, owing to the
mobility of the latter substance, and hence, in spite of the great
activity of birds, no great accelerations are imparted to the body
— the locomotion is not impulsive. The conditions affecting
terrestrial locomotion are exceedingly unlike. Here, as before
remarked, the powers of resistance possessed by the substances of
oi-ganism and substratum both being of high degree, reaction
between the two is correspondingly great. In Reptilia and
(excepting the saltatory Anura referred to below) terrestrial
Amphibia, however, the forces involved in locomotion are not
conspicuous for their intensity, owing both to the fact that the
period of upheaval is prolonged (no sudden impact occurring
between the limbs and the earth) and the small amount of eleva-
tion effected by the action of the limbs, these two features
resulting from the conformation of the body and the general
inactivity of the animal. In Mammalia, on the other hand, the
period occupied by contact of the limbs with the earth is
extremely biief and the height of elevation considerable ; and
hence, though the angulation of the limbs tends to diminish
concussion, the organisation is subject to the intense strains and
stresses resulting from the enormous forces generated during
locomotion. Consider the gallop of a typical Ungulate or
Carniv'ore. The elongated trunk possessing two pairs of limbs,
each pair being in an opposite phase of motion compared with
the other, it follows that the two halves of the trunk will alter-
nately be upheaved through a considei-able distance on contact
with the earth being made by their respective paii-s of limbs, and
depressed in the interval which exists between successive contacts.
That is, the pendulous swing of each pair of extremities is accom-
panied by an elevatoiy impulse at the centi'e of the ai-c each
describes, and these impulses effect the upheavals of the respective
halves of the body through the distances wdiich they have fallen
in the time required for each pair of limbs to describe twice the
length of its ajDpropriate path. Oonsideiing the mode of action
of either pair of limbs, and beginning at the horizontal stage of
the trunk's position when the half of the body is being depressed
under the influence of gravit}^, this continues to descend until the

1903.] TKAxsrosixiox of mammaliax testks. 323

limbs make contact with tlie earth, when, as alreail}' descri^jed,
the entire portion of the trunk is suddenly u])heaved with con-
siderable force. If the horizontal velocity could be reckoned as
constant, the only accelerations which the trunk would acquire
would be in a vertical line ; but since the hoi-izontal velocity is a
quantity which, owing to the i-esistance of the air, is continxially
decreasing, the upward acceleration suddenly imparted to the
trunk at each contact of the limbs with the earth has to be com-
pounded with a sudden forward accelei-ation in order to maintain
the pace, and lience the resultant acceleration which the half of
the body acquires at eveiy upheaval is in an apu-ard i\\\(\ for tear d
direction. It is most essential to I'ecognise this fact that the body
as a whole has periodically imparted to it large accelei'ations in
these two directions. As is obvious, the upwai'd and horizontal
velocities due to these accelerations become respectively nei^tralised
by gravity and atmospheric resistance at the middle and enrl of
the interval which exists between successive contacts of the limbs
with the earth. From the instant when the maximum elevation
of the half of the trunk is attained, i.e. when the upward velocity
is completely neutralised, until contact again occurs, the body
continues to acquire a downward acceleiution due to gravity.
But on contact taking place, the lai-ge downward velocity due to
this acceleration is suddenly destroyed, and as suddenly exchanged
for an equal upward velocity, in the biief instant of time occupied
by the impact of limbs and earth. Whence it is evident that the
degr-ee of upwaixl accelei'ation imparted to the half of the body
during the sudden loss and gain of downward and upward velocities
must be considerable. It isnoteworthy that whereas each /br^t•<:«•f^
acceleration, due to either pair of limbs, is imparted to the body
as a whole, each upioard, acceleration is solely imparted to that
poition of the trunk to which the pair of limbs causing the
elevation belongs ; and hence the attachments of those oi-gans
situated in the median portion of the trunk are evidently not so
subject to downward strains as in the case of organs placed more
tei-minally. Tlius it is important to notice, with regai'd to what
follows, that the primitive pre-mammalian portion of the testes is
decidedly posterior (see diagram, p. 334), and that in consequence
the motion of the hind portion of the body is the factor of most
importance in the consideration of the present problem.

From the foregoing, it is sutHcienth'- manifest that at each
elevation of the body in maunuals a considerable force is imparted
to the organisation as a whole (and by necessary implication to its
constituent parts) in an upward and foi'waixl direction — as was to
be proved. If the constituent pai'ts of the body be considered, it
Avill be evident that such a force woi'ks differential effects among
them. For these organs, lai'gely diflering among themselves in
regard to mass, and the same force being incident on all, it follows
that the individual reactions are different, and hence there exists
a tendency to segregation of those organ;; of greater mass from
those of less. Moreover, oi-gans differ inter se in their relation

324 MR. W. WOODLAND ON THE [Apr. 21,

to the rest of the body. If an organ be imbedded or otherv/ise
firmly affixed to the main bulk, it Avill on account of its large area
of attachment and close apposition readily share both the eleva-
tions and depressions incurred in locomotion — the strains and
stresses are distributed in space and time, and hence their inten-
sity is lesseiied ; if, on the other hand, a massive organ be merely
suspended fi'om the main mass of the body, such a condition
subjects the attachment to the influence of severe strains periodi-
cally recurring, the direction of Avhich approaches that of the
resultant acceleration of the trunk during its impulsive elevation.
The great severity of these strains on the attachment of an organ
thus disposed will be more fully realised if it be remembered
that, on the sudden elevation of the body, the organ possesses the
downward velocity due to the previous depression — a momentum
which in being destroyed throw^s additional strain on the means
of support. As will be seen, the preceding remarks apply in an
eminent degree to the testes.

The Genitalia of Mammalia.

The reproductive organs of the Mammalia are relatively small
bodies of great density and, in the case of the males, compactness,
and they primitively occupy a position similar to that found in
the lower Vertebrata. It is, however, characteristic of the majority
of the Mammalia that in the course of development the testes
forsake their primitive lodgment, and migrate posteriorly and
ventrally to the terminal periphery, where they protrude at the
surface of the body- wall. This protrusion constitutes the scrotum,
in the wide sense of the term, which may vary in character from
that of a pair of small slightly-elevated areas to that of a capacious
oval pedunculated sac. Apparently in man alone the section of
the ccelom contained within the scrotum becomes completely
separated oft' from the main cavity ; in other mammals, com-
nivmication is retained by means of the inguinal canal, which,
however, is usually narrow, so negativing return of the testes to
the main body-cavity. This feature in the case of active
mammals possessing large testes is very important, since if the
unattached organs were permitted to return to the main cavity,
the forces to which they would be exposed during locomotion
Avould doubtless be exceedingly detrimental — a malinfluence which
both the narrowing of the inguinal canal and the possible tractive
function of the gubernaculum during development safeguard
against. In fact, contrary to the usual supposition, the internal
and not the external situation of the testes is the source of
danger. It will also be observed that in small animals which do
not possess nai'row inguinal canals, there is no such dire necessity
for the resti'iction of the testes to the scrotal cavity, both on
account of the smaller size of the organs and the usually less
intense forces to which the testes are subjected. Thus, para-
doxical as it may seem, the increased protrusion of the testes

1903.] TRAXSrOSITlO.X OF MAMMALIAN TESTES. 325

beyon?l the surfcxce of the bo;ly-wall, besiles being a necessary
result of more massive oi-gans and the incidence of more powerful
forces, is in itself a further means of preservation. In physical
character, the testes, as already mentioned, are definite bodies of
concentrated form, suspended dorsally in the body-cavity by the
thin mesorchial membrane. Also investigation shows that, with
the exception of the heart, the testes are the densest oi-gans in
the vertebi'ate body, and hence, as they possess appreciable volume,
their mass is also considei'able. Moreover, the testes (and ovaiies)
are distinctly separated from the remainder of the body, this
separation (lending additional facility to transposition) arising
out of the fact that the gonads ai-e fi-om their veiy nature indi-
vidualised, reproduction essentially consisting of the separating
oil" of a portion of the oi'ganism ; and hence the body prinuuily
serves as a mere carrier of these organs, which, unlike the
kidneys e. g., bear no relation to the economy of the animal.
From this it follows that the transposition of the testes is of no
concern to the rest of the organism, since the process can cause
no derangement of function in other poi'tions of the body ; and
hence the testes, differing from all other organs in this respect,
do not, under oi'dinary conditions, reqidre to be retained in their
])rimitive position by the special development of fixative structures.
Thus, as regai'ds their definiteness and concentration of form,
their means of suspension, their superior density, and their
structural and physiological separateness fi'om the i-est of the
organism, the testes fully coaiform to the above-specified con-
ditions favourable to transjDOsition. The transposition of the
testes occuning under conditions which permit the descent of
these organs alone (and the testes alone have descended), and a
cause capable of effecting this transposition solely existing in the
Mammalia (in which group descent has alone occurred), it is
probable that the latter phenomenon is the cause of the former. I
hold that, in the majority of the Mammalia, the testis attachment
has throughout tlie histoiy of the race been constantly subjected
to severe strains consequent on the character nnd conditions of
mammalian locomotion, and that on account of the resulting
disruption or distension of the mesorchium, the testis has migrated
in a postero-ventral line {i. e. in an opposite direction to the
forward and upward accelerations imparted), coming to lie at the
terminal periphery of the body- wall and forming the scrotal pro-
trusion. Just as when a man runs, a weiglit in his coat-pocket
will periodically " drag " and ultimately wear a hole in the lining
by constant distension, so the testis of mammals has responded to
like forces resulting in " descent."

I now proceed to consider the genital orgnns of the Mammalia
as a Avhole, the general conformation of which amply confirms
the foregoing conclusion, tending to show that the testes have,
in evei-y case, reacted in a degree proportional to the forces
concerned. Indeed, the general correspondence between situation
of the testes and grade of impulsiveness displayed in the various

326 MR. W. WOOBLAND ON THE [^^V^'- 21,

oi'dei's and families of the Mammalia affords such conclusive
evidence as to the causal relation subsisting between the two
that a systematic review of this evidence is well called for. That
the coi'i'espondence is not absolute, however — that there exist
instances of the concurrence of scrotal testes with sluggish habits
— is without doubt mainly, if not wholly, to be explained by
phylogenetic considerations. For it must be remembered that
once the descent of the testes is inherited, the trait is a constant
one (unless acquired sluggishness is able in time to produce efTect,
which is doubtful), and remains so in successive generations
whatever varied habits may be assumed, unless the bionomic
aspect so changes that natural selection cancels the " plus varia-
tions," so affording an ascendancy to reversionaiy factors. That
such has been the case in several instances will be illustrated
below.

In the Monotremes the activity (impulsiveness) is of a veiy
low degree. Ornithoi^hynchus is " aquatic in its habits, passing
most of its time in the water or close to the margin of lakes and
streams." The " body is rather long, compact, and almost every-
whei"e of the same thickness. It rests on short, massive legs

so shoi't that the animal in walking or running actually

drags its body along the ground." Echidna is " usually found in

I'ocky districts, and more especially in the mountains and

is mainly of nocturnal habits." It is described as " indolent."
The testes in both instances are " abdominal in position through-
out life," and afford the only example in mammals of a disposition
of these organs anterior to the kidneys. The low status of these
animals is well known.

In the Marsupials " the testes are always contained in a scrotum,
which is susjoended by a narrow pedicle to the abdomen in front
of the penis." If it is permissible to assume that the ancestor of
the marsupials was kangaroo-like, the pre-penial position of the
testes may perhaps be attributed to the peculiar mode of loco-
motion characterising this animal. Foi- the rapid locomotion of a
Kangaroo consists of a series of leaps, and such leaps would involve,
as in ordinary mammals, a series of antero- dorsal tensions on the
suspensory membranes of the testes ; but since the bodily depres-
sions ai-e mai'ked by a more prolonged termination, as compared
with those involved in the ordinary gallop, the testes would in
addition tend to be thrown' forward, and these two factors in
conjunction have possibly led to the peculiar position of the testes
found in marsupials. On no othei- theory than that of descent
from an ancestor characterised by jjre-penial testes can the peculiar
genital conformation of the variously habited marsupials be corre-
lated Avith special bionomic conditions ; and hence, on this ground
alone, it is possible, as implied above, that the ancestor of the
Marsupialia was of the type of the Macropodidse, the characteristic
Saltatory progression originating the pre-penial situation of the
scrotum in tlie manner indicated.

The Edentata repi'esent anothei- ancient gi'oup. Their activity

1903.] TRyVXSrOSITIOX OF MAMMALIAN TKSTES. 327

in genei'.'il is siimll. Tlie Brarlypodidse are arboreal in habit,
and are chai'acterised by " habitual sliiggishne,ss," being '• most
inefficient walkers," and in climbing, never leaping from bough
to bough (a form of activity invohing more impulsiveness than
even the gallop, but probably not more than that involved in
saltation). Their testes ai'e placed close to each other, lying on
the rectum between it and the bladder; i. e. are retained in the
abdomen. The Myrmecophagidaj are also not noted for activity,
and their testes are disposed as in the Sloths. The Dasypodidie
ai'e similarly inactive, being " harmless and inoffensive," endeax'our-
ing to escape by rapid burrowing. However, it is stated that
" they can run with great rapidity." The testes are abdominal,
lying "above the brim of the pelvis." The Manidje are terrestrial
and burrowing in habit, but some members of the gToup can
climb trees. They are not very active. The testes lie in the
inguinal canal. The OrycteropodicUe (Aard-vai'ks or African
Ant-eatei's) are terrestrial and fossorial. Their testes ai'e
" inguinal, but they appear to descend, at all events temporarily,
into a scrotum." Their phylogenetic position is uncertain. In
the order Sirenia the testes, needless to say, are abdominal.
These animals are " slow and inactive in their movements, mild,
inottensive," browsing at the bottom of watei'.

The piscine locomotion of the Oetacea sutHciently accounts for
the abdominal position of their testes, which organs are placed in
the pi'oximity of the kidneys. Even assuming the tei-restrial
ancestor of the Cetacea to have been charactei-ised by the exter-
nality of the testes (which is improbable considering their existing
localisation), such Avould inevitably have assumed an internal
position consequent on the serious risk of injury involved under
Cetacean conditions.

The Ilodents are comparatively small animals. In habit they
ai-e mostly teri'estrial , but some are arboreal and some natatorial.
" The testes in the rutting-season form projections in the groins,
but (except in tlie Duplicidentata) do not completely leave the
cavity of the abdomen," i. e. the scrotal elevations are not
well-defined. The Duplicidentata compiise the Hares, Rabbits,
and tlie Picas or Tailless Hares, all extremely active animals,
the latter being described as " agile " and as living in crevices
among rocks. The Simplicidentata comprise the remaining
Rodents, the principal families of which are the following : —
The Sciui'ida3 " vary between the two extremes presented by our
ordinary squirrels, the agile climbers, and the sluggish, clumsy
marmots, which live almost entirely underground." Assuming the
statement to be correct that the condition of the testes is similar in
each of these two divisions (which is doubtful), it is evident that
the latter are specialised forms descended from active ancestors.
The locomotion of the Dipodidte resembles that of the marsupial
Maci'opodidjie and insectivoi'an j\lacroscelid;e. " The whole struc-
ture is adapted for jumping, and we iind resemblances in their
btructui'e on the one hand to the kangaroos, and on tlie otlier

328 MR. W. WOODLAND ON XUE [Apr. 2],

hand to the jumping- shrews among the Insectivora." These
animals, unlike the Macropodicla3, move their hind limbs alter-
nately in the pi-ocess of walking. Unfortunately the writer has
been unable to discover detailed information with regard to the
testes in these groups. Other families are the clumsy, thick-set
Hystricidae, the very active Castoridse (Beavers), Myoxidee (Dor-
mice), and Murid^e. During the rutting- season the testes of
Rodents migrate more posteriorly than at normal times ; but this
merely indicates a slight displacement probably due to the periodic
enlargement of the organ, the cremaster muscle eifecting return
by decreasing the capacity of the scrotal emergence. This can be
well seen in the Muridce, where the slightest impact will cause a
dislodgment of the testis.

The order Insectivora comprises "small animals of very

low type ...... belonging to the oldest mammalian stocks

which are generally terrestrial, although rarely of arboreal or aquatic
habits the greater number are cursorial." '• In the sub-
family Centetince, and Chrysocldoris, the testes lie immediately
behind the kidneys, but in others more or less within the pelvis ;
during the rutting- season they become greatly enlarged, forming
protrusions in the inguinal region." The various families of this
group afford good illustr-ations of the correlation of impulbiveness
with degree of testis transposition. As stated above, in the
Centetince or " Crawlers " — the appellation denoting their mode of
locomotion — and Chi-ysocMoris, which is inactive and fossorial,
the testes lie just posterior to the kidneys, i. e. are abdominal.
In the companion subfamily, the Oryzorictince, the two genera
are represented by small animals — Macrogale being mouse-like
with a long tail, and Oryzorictes mole-like in form. The testes
are situated near the urethra. In the Erinaceinm or Hedgehogs
the testes are situated on the " underside of the inguinal canal."
Their " movements are sluggish, their steps almost tottering, their
gait clumsy." On the other hand, in the active Soricidas (Shrews)
and saltatorial Macroscelidte (Jumping-Shrews), the testes project
at the periphery of the perinteum, and in the Solenodontidte, the
feet of which are "formed for running," the "testes are received
into perinatal pouches."

In the order Ohiroptera, the members of which are so highly
specialised for flight, there is " no scrotum, and the testes are
either abdominal or inguinal." " "We find in the low organization
of their brain a proof of their inferior status " — a fact otherwise
implied by the absence of the scrotum, unless a secondary reten-
tion has occurred owing to the danger of partial externality
involved in the conditions of flight and position of the hind limbs.

The order Ungulata is subdivided into the Ungulata vera and
the Subungulata. In the former suborder, needless to say, the
testes of the large majority reside in a well-defined scrotum, cor-
responding to the eminent impulsiveness of the animals. In the
Nasicornia (Rhinoceroses), however, the testes are inguinal, the
tunica vaginalis communicating freely with the body- cavity. As

1903.] THAXSrOSITIOX OF MAMMALIAX TKSTES. 329

is well known, they five " luigo, heavy, clumsy creatures, with
bent legs so short that the belly seen)S almost to drag on the
gronnrl." The Subungulata comprise the Hyracoidea and the
Proboscidea. The ancestry of the mendiers of both these sub-
orders is as yet undecided, their status, decidedly low in the scale,
being obscure. The Hyracoidea are animals of about the size of a
rabbit or somewhat larger, and " resemble small maiinots." They
possess a " short, fat body " with "weak and short feet." "In
most species there is a complete adaptation to a life among the
rocks," by possession of curious clinging habits resembling those of
the Geckos. They are " agile in their sports, but lather lazy where
food is abundant." Owen states that " the testes are abdominal,
below or beyond the kidneys." The low status of the Proboscidea
is shown by several anatomical traits, such e. g. as the possession of
two venie cava? and the structure of the limbs. These animals
have been well described as " peaceable colossi." Their gait is
" pretty slow, though the Colossus can van very fast when once
in full career, but this pace never lasts very long and is always
maintained in a straight line." Owing to their huge size
Elephants never gallop. The most active niendiers are the
" I'ejected males" — such extiu activity thus being neutral so far
as the inheritance of any tendency to testis descent is concerned.
Elephants are vegetable feeders and ai'e " gregarious, generally
inoliensive and even timid, fond of shade and solitude and the
neighbourhood of water." The testes are permanently abdominal
—a fact explicable by habits, though phylogeny alone can aflbrd
a complete solution. Here, as possibly also in the case of the
Nasicornia, it may be pointed out that the huge size of these
animals itself implies inactive ancestors, for, according to one of
the conditions of gTOwth enumerated by Spencer, great activity is
antagonistic to increase by bulk, and the occuirence of the latter
negatives the past existence of the foimer.

The order Cai-nivor.a is subdivided into the Fissipedia and the
Pinnipedia. The mend^ei's of the former group are all exceedingly
active animals of large size, and, together A\ith the Ungulata, include
the swiftest of terrestrial animals. Their testes, needless to say,
reside in a well-defined sciotum. The Pinnipedia aftbrd an inter-
esting illustration of the secondaiy operation of natural selection.
In the Otariida; the " hind feet are turned forwai'ds under the body,
and aid in supporting and moving the trunk as in ordinary

mammals They spend more time on shore, and range

inland to a greater distance than the true seals." In the Otariidje
the testes are " suspended in a distinct external scrotum" '. On
the other hand, in the Phocida^, "the hind limbs are directed so
far backAvards that they continue the hoiizontal direction of the

vertebral column They move on land only with difficulty

by fixing themselves with their flippers in front and pulling up
their hinder parts, then drawing their bodies up into a curve and

' Owen states that the bcrotuni is uot distinct.

330 MR. W. AYOODLAND ON THE [Apr. 21,

throwing their front parts again forwards. They drag their belly
along the earth, show little suppleness, and soon become tired."
There is no scrotum, the testes being abdominal and " imbedded
in areolar tissue." Thus, as befoi'e maintained, the embryonic
ti'ansposition of the testes tends to be checked if external conditions
become so changed as to render complete descent of these organs
disadvantageous to the animal, such being obviously the case in
the Phocidae, which have become further modified for an aquatic
existence than the Otariidfe. It may be objected that ordinary
mammals are subject, though in a less degree, to a like drawback
— that their testes are also exposed to considerable danger, and that
on the hypothesis descent ought to have been negatived by natural
selection, i. e. by the development of ligaments or supporting
areolar tissue. Incidentally noting that the actual facts prove
that no such danger exists, it may be observed that the testes of
a typical mammal are very efficiently screened, not only laterally
by their position between the broad thighs of the hind limbs, but
also posteriorly by the tail ; and they are well preserved from
contact with suiTOunding objects by the elevation of the body
upon its limbs, the case being otherwise in Phocidge, Cetacea, and
lower animals.

The ancestral history and arboreal or other habits of the
Primates constitute sufficient warrant for the conspicuous exter-
nality of their testes. That active arboreal habits involve
impulsiveness of the highest degree, is sufficiently manifest on
contemplating the movements of any of the ordinary monkeys,
moi-e especially in the case of the Gibbons.

Thus in a review of the Mammalia, we encounter a considerable
mass of evidence testifying to the validity of the theory here
advocated, and more might be added. In the Monotremata,
Sirenia, Cetacea, most Edentata, Hyracoidea, Proboscidea, and
Phocidpe, conditions prevail, or have prevailed, negativing the
descent of the testes ; and these conditions have either consisted
of the absence of that type of tei-restrial locomotion which has
been the sole cause of the ti'ansposition of these organs, or of
secondary factors which have either negatived the operation of
the primary agency, or effected a reversion of the pre-existing
effect of the same.

Since the tiansposition of the testes is mainly due to their mass,
and definiteness and concentration of form — their means of
suspension and physiological separateness fi'om the body merely
constituting conditions to the transposition — it follows that the
small and diff"use ovaries of the Mammalia will not respond in
any degree to the forces incident on the body, the magnitude of
the strains on an attachment obviously being proportional to the
mass of the organ attached. But apart from the smallness of
mass possessed by the ovaries, thei'e exists an important reason
for their retention at or nea,r the primitive position in the body-
cavity. This reason is the necessity of the proximity of the ovary
to the oviducal aperture for the conservation of the ova, the ovary

1903.] TRAXSrOSITIOX OF MAMMALIAN TESTES. 331

and tlie oviduct usually pnssossinGj no aniitoniical connection.
How inipoitant this is may not only be infeired from a jjriori
considerations, but also from the discovery of the many structural
devices ada.pte<l to this end. Thus thei'e occuis the " develop-
ment of special folds of the peiitoneuni, which practically ensure
the passage of the ova into the oviduct when they are extrudetl
from the ovaiies. The oviduct, moreovei', has a large and
finibi'iated mouth, called in human anatomy the ' morsus diaboli.'
This almost wraps round the ovary, and thus prevents the ova
from straying in the wrong direction. Moreo\"er, the ovary itself
is often so ai-ranged that it can easily be withdrawn into a pocket
of the peritoneum, from which the obvious exit is by the gajiing
mouth of the oviduct. This disposition of the generative parts is
still further modified in a few animals, such as the liat and the
Kinkajou. In these animals the mouth of the oviduct actually
opens into the inteiior of a closed chamber which contains the
ovaiy " {Beddard). Thus there exists ample reason for the
retentive ligaments usually associated with the mammalian ovary
(see Appendix).

Tlie Genitalia and Conditions of Locomotion in Ichthyojjsida
and Saitro'jjsida.

If, as we have observed, the conditions of manimaJian loco-
motion alone subject the oi'ganisation to those concussiv e influences
wdiich have efiected, among other changes, the transposition of
the testes ; and if, as we have also seen, only the highest mani-
festations of terresti'ial activity are capable of producing complete
testicular descent, then we may be certain that neither in the
relatively inactive tei-restrial Reptilia, the active aerial Aves, nor
the aquatic Pisces will a like phenomenon occur. In Heptilia,
though Ophidia and many Lacertilia are capable of bi-ief spasms
of great activity, the total impulsiveness is very small and of low
degree. This is not only due to general passivity, but also to the
fact that " the body of a reptile is, as it wei-e, slung between its
lin\bs like the body of an eighteenth centuiy cliaiiot ])etween its
four wheels," with resulting impei-fection of the " i-elations of the
limbs to the tiunk from the point of view of a terrestiial
creatuie" [Beddard). Apart from the absence of the causal
conditions, the depi'ession of the trunk and consequent liability
of injury to the testes would in itself have negatived descent
{cf. Phocidje above).

In the Anura, the only terrestrial amphibia whic^h concern us,
adoption of terrestrial habits, resulting in a higher degree of
impulsiveness, has as visual for its concomitants structural con-
centration and increased retentivity or transposition. Here the
testes have assumed a " full oval form, compact and undivided :
they are situated, as shown in the Frog, on the ventral side of
the anterior half of the kidneys" [Owen). That the testes
have slightly descended in correspondence with the saltatory

332 MR. W. WOODLAND ON THE [Apr. 21,

progression, is sliown on comparison of two such forms as Triton
and the Frog (see diagram below, p. 334). If, as we admit, anuran
progression involves on a small scale those impulses to which we
attribute the descent of the mammalian testes \ wh}^ have not
these organs migrated more than they have in tliis instance ? In
the first place, it may be contended that the total impulsiveness of
a frog, though quite appreciable, is yet of small value as compared
with tha,t of a mammal of the same dimensions, say the Common
Mouse. Secondly, the physiological and structural relations
obtaining between the testes and the kidneys afford sufficient
grounds for the special retention of the former, and, moreover,
the firm attachment of the testis to the kidney is shielded from the
abruptness of incident strains by the slight mobility of the lattei'
organ. Finally, in addition to the above considerations, it is
probable that the descent of the testes would here be prohibited
in consequence of the depression of the hinder portion of the
trunk and resulting pi'oximity to the external surface. However,
that such a cause for retention might be operative, it would
obviously necessitate transposition to an extent that enabled the
organs to experience adverse influences, and it is questionable as
to whether they have travelled so far.

In the typical Aves, as in Mammalia, the expenditure of energy
is very great, but in the former, as also in Pisces, aquatic
Amphibia, &c., the bionomic conditions negative impulsiveness of
locomotion, and their organisation is not subject to the kinetic
influences which affect terrestrial animals.

One noteworthy subject is the large increase in volume of the
testes during the brief courting-season : thus Owen provides a
figure illustrating the periodic enlargement of these organs in the
Common Sparrow, showing that their size ranges from that of a
pin's head to that of a fair-sized marble, half an inch in diameter;
and this same periodic increase of volume, though not so marked,
occurs in sundry orders of the Mammalia, e. g. in Rodents and
Insectivores. It is easily comprehensible that if an animal's
activity, however impulsive, were largely restricted to the inter-
breeding periods, the testes, possessing small mass during that
time, would exhibit less tendency to displacement ; but as yet it
is not possible to decide whether this factor has any significance.

Connected with a consideration of the position of the testes in
birds, is the important case involved by the habits of the
Ostriches and their allies. The considerable speed attained by
these birds, comparable indeed with that of the swiftest mammals,
must involve a certain amount of iropulsiveness ; and such being
the case, why have not the testes responded by descent ? For the
situation of the testes in Ostriches is normally avian, being " p»laced
above and a little external to the kidneys," which latter structui'cs

are " elongated, flattened, glandular m.asses tying deeply

seated, and extending from the posterioi- edge of the diaphragm

1 " The great Bull-Frog may clear six feet at a leap, and repeat them so rapidly
as to escape a pursuer, unless chased at a great distance from the water " (Owen).

1903.] TRAXSPOSITION OF MAMMALIAN' TESTES. 333

to the .interior extremity of the pelvic cavity " [Macalister). It
is obvious that noii-clescent cannot be ascriljed to the serious risk
of injury that externality would involve (for, unlike as in the
maniiuals, the position of the proportionately thin legs and the
absence of a tail ensure no protection from external influences),
since such an exjjlanation is negatived by the present position of
the testes. However, if the several chai'acteristics of the mam-
malian gallop be called to mind, it is easy to aftbrd an explanation
of this apparently anomalous instance of tlie non-transposition of
the testes. For, as I shall show, the bipedal progression of the
Ostrich does not involve impulsive upheavals of the body to any-
thing like the same extent as in the gallop of a mammal. The
great upward impulses associated with the upheavals of the mam-
malian body are in large part due to the elongation of tlie truidv,
for the depression of one half of the body is largely accelerated by
the impetus imparted to it on the sudden elevation of the other
half. That is, the upheaval of one half of the body x-otates the
whole lengtli of the trunk about its centre of gravity, and so
causes tlie other half to descend with greater acceleration tlinn
it would do under the sole influence of gravity. Plence the
marked intensity of the contacts of the limbs with the earth,
which have not only to destroy the downward velocity due
to gravity, and the additional downward velocity due to this
lever-like action of the trunk as a whole, but have also to
impart an upward velocity equal in amount to the two it has
neutralised. In the Ostrich, on the other hand, there is no elon-
gated trunk possessing two pairs of limbs, and hence there can be
none of that lever-action just described. Moreover, the individual
elevations of the body are small compared with those of mammals,
owing to the alternate action of the limbs (c/. the Kangaroo, in
which the action of the limbs is not alternate, and in which the
elevations are consequently great). And lastly, the gliding
motion adopted by the Ostrich in running is yet anottier factor
tending to diminish the intensity of the involved concussions.
Hence for these three reasons the individual impulsive accelera-
tions imparted to the body of the Ostrich are very small compared
with those which exist in the case of mammals ; and therefore it
can easily be understood how it is that, despite the terrestrial
locomotion and the perpendicular limbs, the testes have not
descended in this instance. It is also possible that the posterior
extension of the sternum and ribs in the struthious trunk,
leading to tight packing of the viscera, so serves as a support for
the testes (and other organs), but whether this is the case can
only be determined by I'eferring to the facts.

In conclusion, it will not be out of place to here give a brief
summary of the entire argument concerning the cause of the
descent of the testes, since it will show how strong is the proba-
bility of the theory advanced. "We have seen that in mammals
alone there exists a meclianical cause competent to cft'ect trans-
position. If any doubt remains in the mind of the reader as to

334

MR. W. WOODLAND ON THE

[Apr. 21,

the competency of this alleged cause, contemplation of the actual
locomotion of a horse or dog, or reference to the works of Marey,
Pettigrew, and others will soon dispel it. Further, I have
shown that the degree of transposition of the testes corresponds

Text-fig. 57.

CLUPEA

SCYLLIUM

TRITON

RANA

LACERTA"

COLUMBA

MUS

FEUS

Diagram illustrating Testis Descent.

This diagram, based on a series of measurements, illustrates the positions
and lengths of the testes relatively to the length of the trunk proper (minus tail) in
Chipea, Scyllium, Triton, Hana, Lacerta (muraJis), Columha, the common mouse,
and the cat, all measurements being reduced to one scale. The vertical sides of
the rectangle represent the length of the animal to which the other measurements
are proportional, and the dots the approximate positions of the centres of gravity of
the testes. Such a figure well exhibits the concentration and transposition corre-
sponding to higher grades of impulsiveness. To fully illustrate the relationship a
similar diagram drawn up from several hundred dissections is necessary, together
with the average weight of the testis and intensity of the reactions incurred in
locomotion, appended in the case of each animal.

1903.] TRAXSPOSITIOX OF MAMMALIAN TESTES. 335

to the intensity of the forces concerned, ami have tluis sup})lied
the inducfcive evidence required to verify the conckision arrived
at on a priori grounds. It h.is also been pointed out that
transposition of tlie testes can, from the nature of the function
of these oi-gans, the mode of emission of their genital products,
and the absence of adverse external influences, neither cause
derangements in the individual economy nor impair the fertility
of the race ; and that in consequence of a similai- change of
position in the case of other oi'gans inevitably entailing one or
other of these penalties, the testes alone have descended. More-
over, the supei-ioi" density, the disposition of the attachment, and
the appropriateness lioth as regards volume and tlefiniteness and
concentration of form for change of position, appertaining to the
testes, lend greater additional facilities to their transposition than
in the case of othei- organs not so characterised. Hence we can
not only show why the testes have descended, but also why they
alone have thus responded to the incident forces — other organs,
for one or more of the reasons supplied, requiring to be maintained
in position l)y the special retentive structures developed to tlia,t
end. Combining with these facts the additional evidence derived
from a study of mammalian anatomy, showing that imiDulsive
locomotion is not alone responsible for the descent of the testes,
but that it has also been the prime cause of the evolution of the
diaphragm, the metanephi-ic kidney, and other minoi- anatomical
features, we see still moi-e clearly " how strong is the pi-obability
of the theory advanced."

Other Ulustratious of the Relationshljy between Visceral Con-
formation and Impulsiveness : the Evoliction of the Diaphragin
aiid the Metanephric Kidney.

Althoiigh difiering from the reproductive elements of the body
as regards their economic i-elations to the rest of the organism,
the non-repi'oductive organs, possessing like mateiial propei'ties,
may be expected to portiuy simUai- structui-al chai'acteiistics in
i-elation to locomotor impulsiveness. A consideration of such will
not only prove of utility as affoi-ding indirect confirmation of
the foregoing, but will pi-o\dde opportunity for tendei-ing similar
explanations in the case of other anatomical featui-es. The con-
stant intei'i'elations of position obtaining between the various
oi'gans in the vei-tebi-ate body a.i-e in eveiy case easily accounted
for, either on mechanical or physiological grounds, and it is
needless to discuss every oi-gan fi-om these points of view ; it will
suffice if we treat of two or thi-ee structural phenomena, the
interprets! tion of which is not quite so self-evident or well
known.

The complete descent of the testes and the pi'esence of a fully-
developed diaphragm both being conmion chai-acteristics of the
Mammalia, it is possil)le that the cause to which the foi'mer is
attributable may also suffice to explain the evolution of the lattei-,

Proc. Zool. See— 1903, Vol. I. No. XXII. 22

336 MR. W. WOODLAND ON THE [-^V^'- ^Ij

and enquiry confirms the supposition. Though, in the above
account of mammalian locomotion, most stress has been laid upon
the impulsive elevations of the body and resulting strains on the
attachments of organs, yet it must not on that account be inferred
that on descent of either half of the trunk the shock consequent
on contact of the limbs with the earth (equal in intensity to the
elevatory impulse) is negligible. (See above in case of Kangaroo
in which it is taken into account.) Such shocks have the effect
of causing those viscera which are closely adherent to the mass of
the body either to exert a considerable pressure on strvictures
anteiior to them or to be dislodged from their normal position ;
and if in either case such a result is to be avoided, organs of
support must be developed. In mammals, the lungs together with
the heart occupy the anterior portion of the coelomic cavity, and
behind these are situated the liver, stomach, and intestines, these
together constituting a considerable mass. These massive organs,,
unless prevented, would on each contact of the limbs with the
earth exert great pressure on the fragile compressible lungs
immediately anterior to them. Hence, in order to obviate
ensuing derangements, we find in all mammals, and in many other
terrestrial animals, a stout partition separating the cavity of the
liver and gut from the cavity of the lungs — a structure which can
only have been originated by natural selection. Evidence sup-
poi'ting the conclusion that the diaphi'agm arose as an adaptation
to the forward pressure of the liver and gastric mass, is not only to
be found in the fact that the diaphragm is convex anteriorly, but
also in that the convexity is, cceteris jmribus, proportional to the
impulsiveness of the animal's activity and to the mass of the
liver and gut. In herbivores the mass of the gut is greater than
in carnivores ; on the other hand, "carnivorous (fat-eating) animals
generally possess a larger liver than hevhixores " {Wiedersheim),
so that these respective characters tend to defeat comparison \
But in the Perissodactyla we find a fairly large liver (considerably
larger than in Artiodactyla), a long herbivoran gut, and a high
degree of impulsiveness, the combination of which on our hypo-
thesis should be correlated with a higlily convex diaphragm — an
inference which proves correct. Owen remarks that " in the
perissodactyle Ungulates, in which the movable ribs are numerou.s
and continued to near the pelvis, the diaplxragm is also extensive
and much arched towards the thorax " — indicating that the con-
vexity of the diaphragm is markedly above the normal. And,
doubtless, other illustrations are adducible. Additional evidence
as to the primary function of the diaphragm is, moi'eover, afforded
by the case of the Struthiones (and allies) wliich have adopted
the mammalian mode of locomotion. In these birds there exists
a " well-marked diaphragm forming a pai-tition which divides the
thoracic cavity into two parts, one posterior and small containing
the lungs, and the other anterior and large containing the heart

1 The small capacity of the herbivoran thoracic cavity as compared with that of
the carnivora must be remembered in this connection.

1903.] TRANSPOSITION OF MAMMALIAN TESTES. 337

and liver. It is a fibi-ous membrane, concaA-e forwainls, -with a
muscular attachment at eithei* side to the ribs and intei-costal

tissues, which it joins in about the middle of their course

The pleural cavity is closed above and below by the fil^rous dia-
phi-agm becoming blended with the first and last i-ibs. The
anterior thoracic cavity, which contains the pericardium-coated
heart in its upper ]jart, entirely independent of the pleui'al
ca\'ity, is di^•ided into two by ;i dense fibi'ous meinbiune
which springs from two vertebral crura, much as the human
diaphragm, and extends above the line to join the sternum along
the border which articulates witli tlie ribs, leaving the heart entirely
in fi'ont of it ; its concavity is directed downwai-ds and foi-wai'ds,
and it is separated from the diaphragm propei' ' )y veiy large aii'-cells.
The liver is completely separated from the al^dominal cavity by a
fibrous membrane, so that when the included viscei'a are removed,
it is not at all bi-ought into view. The mesentery is very dense
and strong " {Garrod ((• Dancin). A comparison of the struthious
and mammalian diaphragms afibrds furthei- confirmation. In the
Ostrich, owing to tlie anomalous position of the avian lungs (their
close application to the costal skeleton ensui'ing the intactness of
the air-cell connections), the liver has practically assumed their
role, as i-egards conformation, in relation to the heart and dia-
phi-agm. Hence the diaphi'agm is concave anteriorly, consequent
on the necessity for the retention of the liver. In the mammal,
on the other hand, the lungs are of large volume and freely sus-
pended in the ventral portion of the thoi-ax, enveloping the heart.
If the liver wei'e anterior to the diaphi^igm it would, as before
remai-ked, during locomotion pei-iodically exei't gi'eat pressui'e on
the lungs ; hence the diaphragm here becomes the foi-ward sup-
port of the liver and other alimentary viscera. The envelopment
of the heart by the liver in the Ostrich possibly affords an illus-
tration of that alternative to the formation of a diaphragm above
referred to.

From these and other considerations, we can more precisely
intei-pi-et the anterior convexity of the diaphi'agm, which is due
to two causes : the forward pressure in the median line of the
large abdominal viscera, and the backward lateivil extension of
the capacious lungs. Evidence as to the cooperation of this last
factor is affoixled l^y such cases as that of the Manatee, in which,
on account of its subaqueous habits, unusuall}' developed organs
of respiration ai'e leijuired, it also lieing essential that their ex-
tension should be dorsal in older to ensure the ventral situation
of the centre of gravity of the liody. Here, as Mivart remarks,
the diaphragm is so oblique that the thorax " extends backwards
above the whole length of the abdominal viscera" ; and a similai-
conformation is found in Cetacea and perhaps other groups. In
development, the diaphragm is " formed from a couple of septa,
dorsal and venti-al respectively, which arise independently, and
are for some time quite di.stinct from each other " [Marshall) :
from which fact it follows that the part the diaphragm plays in

o38 ME. W. WOODLAND ON THE [^Pl'- 21,

connection with respiration cannot have constituted its primitive
raison cVetre.

Thus, observing the necessity for the development of the dia-
phragm, we find that its variations in disposition and contour
found in the Struthiones and Mammalian orders fully confirm
the conclusion deduced on a priori grounds.

Another organ which, though not peculiar to the Mammalia,
is yet a diagnostic feature of terrestrial vertebi-ates, is the meta-
nephric kidney, and that there is possibly a relationship between
teii-estrial modes of locomotion and the evolution of this organ
I will now endeavour to show.

It is well established that the metanephros is a development
of the definitive mesonephi'os which has lost its nephrostomata,
acquired a separate duct, and become more or less concentrated
in form. It has been pointed out above that concentration
of structure is an essential concomitant of that increase of
impulsiveness which results from the increased activity of the
animal under conditions which involve marked reactions between
the body and the medium or substratum. That this is so in the
case of the kidney, it is only necessary to compare the elongated
mesonephric bodies of Pisces and aquatic Urodeles with the same
organs of the teirestrial Amphibia and the metanephric bodies
of Reptilia, and again to compare these latter with their repre-
sentatives in Mammalia, in which both concentration of structure
and impulsiveness of locomotion attain their maxima. With
i-egard to the loss of nephrostomata, the same relation holds.
In all Pisces, with the exception of certain Elasmobranchii, the
mesonephi'ic bodies I'etain connection with the coelom by means
of the nephrostomial tvibules. Whether the absence of these
fragile structures in Elasmobranchs (their function with regard
to the ccelom perhaps being assumed by the abdominal pores) is
to be attributed to the fact that these fishes either are or are not
descended from the most active members of their class (and it
must be remembei'ed that they are surface forms) is not certain,
but it is possible. In the aquatic Ui-odeles the nephi'ostomes are
present, but in the terrestrial Anura they have completely lost
their connection with the coelom (which lattei-, as in Elasmo-
branchs, has had to discover anothei- means of exit for its waste
prodvicts). The causal relation between disruption of the nephro-
stomata and the adoption of a tei-restrial life is here clearly shown
by the ontogeny. In the development of the frog, the meso-
nephros is at first in communication with the coelom by means of
the nephrostomata, but at the period of metamorphosis this con-
nection is severed, and the kidney, becoming more concentrated,
finally assumes the definite form of the adult structui'e. ISTeedless
to say, the metanephros of Reptilia, Aves, aiid Mammalia is
totally devoid of nephrostomes, that of the last not possessing
them at any stage of development. The division of the meso-
nephros into two poi-tions, one coming into I'clation with the
testes, and the other — the definitive mesonephros — retaining its

1903.] TRA>SrOSTTIOX OF MAMMALIAN TKSTES. 330

uiinaiy function, is evidently a physiological phenomenon, though
the ultimate structural separation of the two parts must largely
he attributed to the influenee of greater activity of the organism.
Also the secondarv development of the definitive mesonephros is
indirectly due to increased activity, since the worlc of the excretion
resultina- from the latter is largely increased. In fact, the whole
ontogenv of the kidney— the successive developments of the pro-,
meso""-, and metanephros in space and time, the diflerences obtaining
between the pro- and mesonephric tulmles. tlie disi.ppearance ot
the pronephros, &c., &c.— can be directly or indirectly attributed
to the increasing activity of the organism in the course of phylo-
genv, as a little thought will show.

" With regard to the position of the kidney— a l)ody of consider-
able mass, and in Amniota similarly disposed to the testes— no
relation between such and the impulsiveness of the animal being
traceable, it is evident that some other factor must be involved
aiid enquiry shows that the localisation of the kidney ls correlate.!
with the total activity of the animal. Excretion being facilitated
h\ proximitv to a vigorous blood-supply, we find that m active
animals the "kidney tends to be situated anteriorly, i. e. nearer the
heart ; on the other hand, close connection with tlie vent to the
exterior is similarly advantageous, and in inactive animals, such
as Reptilia and Urodeles, we find the ki<lneys situated at the
posterior extremity of the coelom.

Tlie kidneys of mammals are retained in position b>- a circuin-
ambient development of areolar tissue, which usually contains
much fat ; support is also contributed by the " tonicity " of the
muscles of the abdominal wall and by the terminal ribs.

In conclusion, I must acknowledge my indelitedness for details
respectinc the habits and anatomy of the mammalian orders to
the following authors and works :' Flower and Lydekker, Owen.
Wiedersheim. Beddard and Gegenbaur, Lydekker's 'Royal
Natural Historv.' 'Mammals' by Yogt & Specht, Schmeil's ' Text-
book of Zoology,' and Allen's & Lloyd's editions of Jardmes
' Naturalist's Library.' .

I also wish to thank Prof. Minchin, Mr. J. T. Cunningham, and
Mr. H. S. Sheltou for kindly reading through the manuscript and
for making several suggestions and criticisms.

Appexdix.
In respect to the non-descent of the ovary in the Mammalia, it
is as well to emphasise one factor ensuring its retention. Assum-
in<v\hat the females of mammals are approximately as active as
the males, it is evident that if the mammalian o^nlm had retained
its volk {i. e. if the mammalian ovai'v possessed the size and mass
of that of the Sauropsida and Monotreines), the retention of the
ovary within the body-cavity by the development of special liga-
ments, &c., would have been, if not impossible, yet extremely
hazardous. And such being the case, it woiUd obviously benefit
the race if the potency of one of these U\o factors were decreased.

340 MR. R. I. POCOCK ON THE [Apr, 21,

Now, under the conditions of mammalian life — conditions in which
survival of the fittest attains within its limited sphere of operation
a maximum degree of efficiency — it is obvious that diminution of
activity wovild be fatal, the speed of mammals being one of the
most important conditions to survival in the struggle for existence
(as is shown by the fact that this trait is so highly developed in
these animals) ; hence, a less degree of activity being prohibited,
any decrease in the mass of the ovary would be of service to the
organism in which it occurred. Natural selection may legitimately
be svTpposed to operate here, since, although it has been clearly
demonstrated that the higher the life of the organism the less
range of application does this principle possess, yet it doubtless
applies in the case of any feature which is of paramount importance,
and modifications concerned with the genital structures must
necessarily possess such importance. It is therefore possible, a,nd
even probable, that the loss of yolk sufi'ered by the mammalian
ovary and the alternative adoption of a placental mode of nutiition
both indirectly result from that same cause of impulsive locomotion
to which we have traced several other features of mammalian
structure.

It may also be worth while to add that many minor features of
mammalian anatomy, the significance of which is usually over-
looked, are only explicable on the assumption that they aie related
to impulsive locomotion. Instances of these minor structures are :
the accumulation of fat at the base of the heai't, the fatty cushion
surrounding the neck of the bladder, the fatty development about
the kidney already noticed, and the various " suspensory ligaments "
and other " fixative organs " referred to above, associated with the
stomach, liver, and other viscera of lara,'e mass.

5. On the Geographical Distribution of Spiders of the
Order Mygalomorphse. By E. I. PocoCK, F.Z.S.

fEeceived March 17, 1903.]

(Text-figures 58-61.)

Part I.

Introductory Remarks upon the Paleontology and the
Means op Dispersal op Spiders.

(«) Summary of the Palfeontological History of Spiders, and
its bearing on the Phenomena of Distribution.

Owing to the enormous chances against the pi'eservation of
fossil Spiders in sedimentary rocks, the palseontological histoiy
of this Order is very imperfect. One or two types have been
discovered in Carboniferous strata of Europe and North Ameiica
[Arthrolycosa and Protolycosa), and also a fairly large number of
specimens from amber and from gypsum and lacustrine deposits
of Oligocene and Miocene age in those countries. But absolutely

1903.] GEOGRAPHICAL DISTRIBUTIOX OF SPIDERS. 341

nothing is known of the forms that inhabited tlie world during
the enormous lapse of time represented by the Mesozoic stiuta,
and nothing except inferentially of the types that occupied the
!southei-n countiies of the world dni-ing Tei-tiary times.

Spiders of the type that lived in the Carboniferous period
succeeded in holding their own in Europe until the Oligocene, and
are represented at the present time by the genus Liphistius, which
is restricted to the Indo- Malayan are;i of the Oriental Region.

Apait from the genus Liphistius, aW existing Spiders, including
the Mygalomorpha^, belong to the group Opisthothehe. Thei'e is no
evidence that this group existed in the Carboniferous period ; but
since most of the Oligocene and j\[iocene fossils belong to existing
families, or sometimes indeed to existing genei-a, it is permissible
to suppose that the OpisthotheljB originated some time during the
Mesozoic epoch, and may, in fact, be coeval with the mammalia.
AMiether any of these hypothetic.d INIesozoic foi-ms sui-vive to the
pi'esent day, it is quite impossible to say. All that palaeontology
allows us to infei- is that during the Teitiaiy pei-iod thei-e was
a rich and vaiied spidei'-population spi-ead over the Northei-n
hemisphere, containing forms that have undergone but little
metamoi'phosis since that date. The existence of jMygalomoi-ph?e
at that time is attested by the discoveiy of one foim referred to
Mygale in the gypsum-beds at Aix, and of anothei', Eoatypus,
in the Eocene strata at Garnet Bay in the 'Isle of Wight. But
since it is impossible to classify these forms with an appi-oach
to cei'tainty in any of the existing families, theii- only value
from the geographical standpoint is the evidence they supply
that the Mygalomorphag had come into being in Tertiaiy times, and
wei'e living in the ISTorthern hemisphere.

The imperfections in our knowledge above alluded to pei-mit
oidy a pi'ovisional acceptance of the theoiies put forwai'd in the
following pages to explain the distributional phenomena of the
Mygalomoi-pha?. But all the available evidence, little enough
though it be, points to the conclusion that the jVIygalomorphse
and the rest of the Opisthothelse appeared iirst in the jSTortheni
hemisphere, and spread thence over the southein countries of
the woi'ld.

{})) jVIeans of Dispei'sal of Spiders, and the importance of the
Mygalomorphaj fiom the Geographical standpoint.

It cannot be claimed that Spiders as a whole ai'e a favourable
group to study from a geographical point of view ; foi-, although
exclusively terrestrial when adult, and, like other fliiihtless animals,
dependent upon continuity of land-sui-faces for migration, a gi'eat
many species are known to have the power, and the instinct to
pi;t it in foi'ce, of dispelling themselves over wide areas by
practising when young the habit of flight, using silk-threads as
aei'ial floats upon which they may be cai-ried long distances before
the wind. This phenomenon is well known, and has given rise to

342 MR. R. I. pococK ON THE [Apr. 21,

the belief in the existence of a ' gossamer ' spider which is supposed
to be the caxise of the fine thi-eads which fall from the air and
carpet the fields with silk at certain times of the year. It is now
known that the ' gossamer ' sjDidei- is a mythical species, and that
species of the most divei'se habits belonging to widely difi"erent
fRmilies are responsible foi' the floating threads. The habit is
practised alike, and, so far as is known, to an equal extent, by
snare-spinning forms belonging to the Argiopida3 and Theridiidfe,
by hunting-spidei'S like the Lycosid?e and Attida?, or by sedentary
species that lurk in flowers, like the Thomisidre.

That this method of locomotion may considerably influence the
distribution of spidei's may be infei'red from the fact that cobwebs
thrown out in this way, and afibrding suppoit to little spiders,
have been found at the tops of our highest buildings, and have
become entangled in the rigging of ships 200 miles from land.

There ai-e reasons for thinking, however, that the habit is for
the most part restricted to phanerozoic diurnal species, namely,
those that hunt their prey or spin their webs in the open ; and
that cryptozoic forms, that live in burrows or under stones or logs
of wood, and that are for the most part nocturnal, do not indulge
in it \

Clearly, thei'efore, these cryptozoic groups, in which the restric-
tions to dispersal are piesumably the same as in other terrestrial
animals which can neither fly nor swim to any distance, have more
value for the establishment of geographical areas than those species
with powers of dispersal analogous to flight.

Owing to the relatively large size and gTeat weight of the newly
hatched young of the Mygalomorphas, coupled with the reduction
in the number of spinning-apjjendages and the greater simiDlicity
of the silk-glands, it seems pi'obable that aerial sailing is not
practised to any gi'cat extent by the members of this suborder ^.
Especially true will this be of the Aviculaiiida?, a family which
contains the largest spiders known of this or any othei' epoch,
with newly-born young rivalling or excelling in size the adults of
many species of the Arachnomorpha?.

Consideiution of these facts, coupled v/ith the impossibility of
dealing in detail, in one paper, with the distribution of all the
genera- of the Araiiese, has led to the selection of the Mygalomcrphse
as the fittest group to illustrate the geographical distiibution of
Spiders in general.

1 Simon states that the Spidei'-fauua of the Sandwich Islands is composed wholly
of species of the former categoiy, with the exception of some few forms which appear
to owe their presence in that Archipelago to human agencj- (' Fanna Hawaiiensis,'
Araneaj, 1902).

" The young of the only known British representative of this group, namely
Atypus, one of the smallest types of Mygalom()rph;\% have been seen to scatter over
small areas Ly this method of" travelling \F. Enock, Tr. Ent. Soc. 1885).

In this connection it is instructive to remark that Att/pvs has a wider distribution
than any other known genus of the suborder, ranging from Ireland and Algeria to
Japan and over the Eastern (P the Western) States of North America, that is to say
across the Northern hemisphere from the eastern to the western shores of the
Atlantic.

1903.] GEOGRAPHICAL DISTRIBUTION OF SPIDERS. 343

Part II.

DlSTRIUUTIOX OF THE FAMILIES, SUBFAMILIES, AXD GeXERA OF

MvGALOMORPHiE, and the evidence thus supplied as to their
Oriijinal Habitat and the Lines of Migration followed in
Dispersal.

Family Diplurid^.

The Dipluiidif are the most widely distributed of all groups of
Mygalomorpha>, heing found practically all over the world to the
south of about the 4Uth parallel of north latitude. The numerous
groups, however, into which the genera fall present some featui'es
in their geographical i-ange of considei-able interest.

1. Subfamily DiPLURix.E.^The genus BrachytheW is met with
in the Mediterranean Region, Central Asia, and the Southern
States of Noi'th Amei'ica. Neai'ly allied to it are Hapalothele fi'om
Madagascai- ; Brachythelisciis from Natal ; Aname, Ixamatus,
Chenistonia, iind JJekana from Australia and Tasmani;i; Fujius
fi'om Centi-al and South America ; Trissothele from Chili ; Lycimis
fi-om the Argentine. South America is also the home of Trechona,
Diplura, Uriichtis, Harmonicon, and Melodeits.

2. Subfamily MACROTHELiXiE. — This subfamily is divisil)le into
four gi'oups.

(ci) The Macrothelfe : — Maarothele occurs in Spain, China,
Burma, Singapoi-e, and Java. Neai-ly i-elated ai'e Phj/xiosduema
fi'om Transcaspia, Stenygrocerciis fi'om New Caledonia and
Queensland, and Porrhothele from New Zealand. Ischnothele
is represented liy species from India, Madagascar, S. it W. Africa,
and Central and South Ameiica ; and Evagrios is known from
S. Africa and Central America, whence it extends into the
Southern States of North America (Idaho).

(/>) The Hexathelfe contain two genera — Hexathele from New
Zealand, and Scotincecus fi-om Chili.

(c) The Atraces contain the ^ewei-A Atrax and Hadronycha, from
Eastern Australia.

(fZ) The Masteri;e, comprising Accola and J/^asteria, wliicJi are
probably identical, occur in Yenezuela, the Philippine Islands, and
Upolu.

The presence of Brachythele in the Mediterranean and Sonoran
areas, and of nearly allied forms in South Africa, Mailagascar, and
all over South America, suggests immigration from the north
into these countries of the Southern hemisphere. On the other
hand, the entii'e absence of i-elated types from the ai-ea lying
between and including India and Austro- Malaysia, and the
reappearance in Atisti-alia of geneia closel}' allied to Brachythele

' This genus h;is also been recorded ft'oui S. Africit, Madagascar, S. America, and
Australia. It is probable, however, that the species referred to it belong to one or
other of the allied 2:enera from these areas.

344

MK, R. I. POCOCK OIn THE

[Apr. 21,

If

fi

ft

s

ns

•r-i.

^

s

•S^

c^

=4H

O

o3

M

0)

;:•'•■•

PI

&i3

'•■.■.■.'•

8

r-^.

<D

...,'.

.g

:•■■.■■.•■■

il

•■:-•

tq

-tj

o

o

c^

§

©

r=3

•+3

=4-1

O

Pi

o

^^^

^

1903.] GEOGRAPHICAL DISTRIBUTIOX OF SriBERS. 345

as well as to the South- American and Afi-o-Mascarene foi-ms, points
equally forcibly to the peopling of Australia from either one or
the otlier, or perliaps both, of the southern continents just
mentioned.

The remaining geneia of Dipluiinaj, namely, IWchona, Diplura,
Harmoiiicon, Jlelodeus, and Cruchiis, all of which ai'e more
specialised tjqjes than BracJu/thele and its allies, probably arose
within their present ni'ea. of (Hstril)ution.

A greiit contrast to the distribution of the Diplurinog is presented
by that of the j\Iaci-otlielin;e.

The occurrence of Macrothele in Spain, Cliina, Burma and Java;
of Porrhothele, which is scarcely separal:)le from it genericnlly, in
New Zealand ; of Phyxioschoima iii the Transcaspian ai-ea ; and of
its near ally, Stenycjrocercas, in Queensland and Kew Caledonia,
suggests a southward nngr:itiou of these types from the northern
provinces of the Old Woi-ld into Australia and New Zealand by
way of China and Indo-iVIalaysia. Similarly Isclinotheh, a more
specialised type than Macrothele, pei'haps descended from the north
by way of India into Madagascai-, South Afi-ica, and crossed
thence into South Ameiica, where with Evayrus, which is also
I'epi-esented in Soutli Afiica, it is the only i-epi-esentative of this
group of Dipkn-ida?. The entii-e absence from the Sonoi-an Region
of foi-ms related to Macrothele, Evagrus, and Ischnothele, is opposed
to the supposition that the two last-mentioned genera had a
noi'thern oiigin in Amei'ica..

The Mastei-i:>5 appear to be degenerate f oi-ms of the Maci'othela?.
They ai'e the smallest of all known Mygalomorphfe, and ai-e
essentially ci-yptozoic or lucifugous, living in cavei'ns or under old
decaying vegetation in the dark, damp forests. Hence we can
only pi-etend to a partial knoAvledge of theii- distiibution, and it
would be rash to draw deductions fi'om the fact of their having
been discovei-ed hitherto only in Venezuela, the Philippines, and
Upolu.

The two genei-a of Ati-aces, Atrax and Hadronyche, confined to
Australia, appear to be Maci"othelina3 specialised for a fossorial
life, with Avhich is con-elated certain features impai'ting to them a
superficial similarity to the Ctenizidte.

The Hexathela?, resembling the Macrothela' except in the
retention of an additional pair of spinning-mamilla?, undoubtedly
a primitive feature, ai-e confined, so far as is knoAvn, to New
Zealand and Chili. There seems no reason to doubt that they
passed fi-om one of these countiies to the other by a southei'u
land-connection.

Family Paratropid.e.

This family, specialised both in sti'ucture and habits, is repre-
sente<l by three genera, Paratrojns, Anisaspis, and Anisaspoides,
confined to the Neotropical Region. Its afiinities are doubtful,
but some primitive genus of Diplurida? allied to Brachythele must

346 MB. R. I. PococK ON THE [Apr. 21,

probably be looked to for its ancestry. There seems no reason to
doubt that it originated in the area it now occupies.

Families ATYPIDiE, BllACHyBOTHRIIDiE, MECICOBOTHRIIDiE.

The genera of Atypidse, two in number, scarcely pass south of
the Equatoi'. Atypus, occurring in the Mediteri-anean Region,
and spreading northwards into Central Europe, beyond the 50th
parallel of north latitude, is the